Content uploaded by John Ndavula
Author content
All content in this area was uploaded by John Ndavula on Jul 22, 2022
Content may be subject to copyright.
Social Media and
Political Campaigns
in Kenya
© 2018 John O. Ndavula
Published by Borderless Press
Phoenixville, Pennsylvania
www.borderlesspress.com
info@borderlesspress.com
Design by Melody Stanford
All rights reserved. No part of this book may be reproduced or transmitted
in any form or by any means, electronic or mechanical, including photocopy, recording,
or any information storage and retrieval system, without prior permission from the
publisher, except by a retriever who may quote brief passages in articles or reviews.
Although every precaution has been taken to verify the accuracy of the information
contained herein, the author and publisher assume no responsibility for any errors or
omissions. No liability is assumed for
damages that may result from the use of information contained within.
Library of Congress Control Number:
ISBN-13:
ISBN-10:
First edition
Printed in the United States of America
To my wife Martha, whose love and
tireless support have made this journey
possible, and to my children, Sasha
and Grant, who make the future seem
so bright. And to my parents, Samson
and Janet, for loving, encouraging, and
believing in me.
i
CONTENTS
CONTENTS
Contents ...............................................i
Foreword ...............................................v
List of abbreviations .....................................vii
Acknowledgements ......................................ix
Introduction ............................................xi
Chapter 1
Communication and Political Campaigns ......1
Political Campaigns. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2
Communication and Political Campaigns .....................4
Changes in the Political Campaign ...........................5
The Pre-modern Period .................................7
The Modern Period .....................................9
Factors Contributing to Transition from the
Pre-modern to the Modern Period ........................11
Towards the Post-modern Period ........................15
Conclusion ..............................................17
ii iii
SOCIAL MEDIA & POLITICAL CAMPAIGNS IN KENYA | John O. Ndavula Contents
Chapter 2
Social Media and Election Campaigns ........19
Social Media .............................................20
Facebook ............................................21
Twitter ...............................................24
Comparing Twitter and Facebook ........................27
Social Media and Election Campaigns .......................28
Adoption of Social Media in the 2013
Election Campaigns in Kenya ...............................30
Type of Social Media Used in the 2013 Election Campaigns ..35
Type of Social Media and Level of Oce ..................36
Conclusion ..............................................40
Chapter 3
Diffusion of Social Media in the 2013
Election Campaigns ......................41
Diffusion of Innovations Theory .............................42
Adopter Characteristics and Social Media Adoption ...........46
Candidate Demographics ...............................46
Awareness ...........................................50
Years of Internet Experience ............................52
Technological Factors and Social Media Adoption .............53
Relative Advantage ....................................53
Complexity ...........................................55
Observability .........................................57
Compatibility .........................................59
Environmental Factors and Social Media Adoption ............60
Level of Oce ........................................60
Competitiveness of the Race ............................62
Constituency Type ....................................62
Voter Demographics ...................................64
Party Aliation .......................................67
Determinants of Social Media Use for
Election Campaigns in Kenya ...............................68
Conclusion ..............................................69
Chapter 4
Techniques for Campaigning on Social Media .85
Techniques of Campaigning on Social Media .................86
Techniques of Informing ................................86
Techniques of Involving ................................88
Techniques of Connecting ..............................88
Techniques of Mobilizing ...............................89
Techniques Employed in the 2013 Election Campaigns .........90
Discussion on Techniques of Using Social Media ..............94
Social Media and Political Marketing ........................96
Conclusion ..............................................98
Chapter 5
The Political Impact of Social Media ........101
Social Media and Election Outcome ........................102
Reaching the Youth ......................................107
Democratic Potentials of Social Media ......................109
Conclusion .............................................113
Chapter 6
Summing Up Social Media and
Election Campaigns .....................115
Looking Forward to the Future .............................116
Conclusion .............................................120
BIBLIOGRAPHY ....................................... 121
Index ............................................... 133
v
FOREWORD
Borderless Press is a Knowledge Activist charitable publishing organi-
zation. We operate with an understanding that knowledge production,
distribution, and consumption have, for the longest time, privileged the
Minority World (Europe and USA) scholars and writers at the expense
of the Majority World (Africa, Asia, and Latin America). Consequent-
ly, our publications are works primarily written by Majority World
scholars and knowledge activists. Like all academic publishers, we are
committed to rigorous peer-review processes, celebrating scholarship
that makes important contributions to academic discourse. Beyond
this tradition, we seek to decolonize knowledge production and dis-
tribution through equality, creativity, and justice. is focus guides us
not only in manuscript selection, but also in the ways we approach our
work and organizational structures. Our authors:
• Are from around the world, and especially from the
Majority World
• Choose the language in which they seek to write and publish
their book
• Collaborate with us on all publishing decisions
e books we publish include a range of authorial voices, from
scholarly to experiential and are focused on justice concerns from
around the world.
Most of our books are published through generous nancial dona-
tions. Postcolonial Networks would like to thank our donors for mak-
ing it possible for us to publish this book.
ank you for supporting our work and the work of our authors.
All inquiries should be addressed to Dr. R.S. Wafula at
drwafula@borderlesspress.com.
vii
LIST OF ABBREVIATIONS
ADEO Awareness and Development Organization
CAPF — Coalition for Accountable Political Financing
EAA — East African Association
FORD — Forum for Restoration of Democracy
ICT — Information and Communication Technology
ICTs — Information and Communication Technologies
IEBC — Independent Electoral and Boundaries Commission
IMC — Integrated Marketing Communication
KANU — Kenya African National Union
KAU — Kenya African Union
KOT — Kenyans On Twitter
ODM — Orange Democratic Movement
PNU — Party of National Unity
PYMK — People You May Know
TNA — e National Alliance
TV — Television
US — United States of America
ix
ACKNOWLEDGEMENTS
is book is a thorough revision of my doctoral thesis. In the pro-
duction of this book, I have beneted from the help of many people
both in Kenya and in the United States. While doing my doctoral work,
I was blessed to have two wonderful supervisors, Prof. Hellen Mberia
and Dr. Michael Kamau. I am indebted to them for their commitment,
guidance, and insight throughout my early years of researching the
material for this book. I am especially grateful to Hellen for believing
in me and my potential and supporting me from my beginnings in the
doctoral program.
During its initial shaping, my work beneted from critical com-
ments and questions raised by Prof. Mark D. Johns, who peer-reviewed
the manuscript. His perceptive comments and critical remarks were
extremely helpful and have led to substantial revisions. I also beneted
from the insightful commentary of Prof. Levi Obonyo, which signi-
cantly strengthened the manuscript.
ere are many others whose support and encouragement I must
acknowledge. I would like to thank Prof. Khaemba Ongeti and Prof.
Lily Mabura who listened to my original research idea and took part in
the initial conversations that gave birth to my thesis. Further gratitude
goes to Dr. Peace Agufana, Jackline Lidubwi, and Dr. Beth Mwelu,
xi
x
SOCIAL MEDIA & POLITICAL CAMPAIGNS IN KENYA | John O. Ndavula
with whom I shared my research plans and who encouraged me along.
While writing this book I had the good fortune to cross paths with
several amiable scholars, including Dr. Sarah Wilder, Dr. R. S. Wafula,
and Dr. Pedro Dos Santos of Luther College. I thank them all for their
contribution to my work. I also thank Prof. Esther Mombo for her
commitment to the success of this project. I sincerely thank Alice Bedi
and Prof. Moses Ochanji for their friendship and generosity in the
United States.
e nal dra of this book was written while in scenic Decorah,
Iowa, USA where I spent three months as a Postcolonial Networks
Scholar-in-Residence. Subsequently, I would like to acknowledge the
support of the institutions that made that possible. I thank Dr. Joseph
Duggan and Dr. R. S. Wafula of Postcolonial Networks for granting
me a writing residency and particularly the latter for his hospitality. I
will remember with gratitude the wonderful welcome I received from
Wafula when I arrived in Decorah. I would like to thank the Dean
of Luther College, Prof. Kevin Kraus, for his faith in the knowledge
activism cause. I also thank the Department of Communication where
I was hosted, and indeed the wider Luther College community. I am
grateful to my employer, St. Paul’s University, for allowing me precious
time o from my administrative and teaching tasks to allow me to
focus on this project.
Many thanks to Judy Neunuebel, who provided the original cover
design. e debt I owe my parents for my academic achievement be-
comes increasingly clear as I get older. Consequently, it is with over-
whelming gratitude that I dedicate this book to my father and mother,
Samson and Janet Atsyaya.
I thank my beloved children, Sasha and Grant, for bearing with me
when the PhD project took away their precious time with Daddy. I
am grateful to my wife, Martha Bedi, for her encouragement, love,
patience, hope, and faith in me throughout graduate school and in
the book writing process. It is to her, my children and family, that I
dedicate this book.
INTRODUCTION
Aer witnessing how the 2007–2008 post-election violence in Kenya
impacted my community, I became acutely aware of the gaps between
the needs of voters and the actions of the political class. I was disturbed
by the question of how voters could possibly elect candidates whose
stance on critical issues they knew little about, and then demonstrate
disappointment when these candidates acted contrary to their wishes.
is led me to start a community-based initiative, Awareness and De-
velopment Organization (ADEO), whose vision is to harness informa-
tion and communication technologies (ICTs) to open conversations
about the implications of political choice on the social and economic
well-being of society. I hoped that narrowing the knowledge gap would
encourage citizens to seek knowledge-based solutions not only to po-
litical issues, but for their economic and social agendas as well. I used
Facebook and Twitter, two well-known platforms, to organize young
professionals, many of whom lived in far-ung towns, to dialogue with
their communities on political, social and economic issues. At the
time, I didn’t know I was engaged in a form of knowledge activism that
would set me on a journey of many years of research into the prospects
of ICTs for political communication.
xii
SOCIAL MEDIA & POLITICAL CAMPAIGNS IN KENYA | John O. Ndavula Introduction
xiii
ICTs in Kenya were rapidly developing and expanding the space for
dialogue. e Kenyan government was committed to developing the
country’s ICT infrastructure as a tool for socio-economic and political
growth as encapsulated in Vision 2030,1 as well in the National ICT
Master Plan 2017.2 Chief among the strategies was the laying of an in-
ternational bre-optic cable that would potentially increase bandwidth
across the country. ese developments, coupled with increasing mo-
bile phone penetration, were positioning Kenya as a hub for techno-
logical development. It is no surprise, therefore, that Kenyans became
the most active users of Facebook in Africa, as well as the fourth most
active users of Twitter in Africa, in just a few years. 3
All these technological and policy developments, coupled with my
work with the larger community, compelled me to research the poten-
tial that social media could have for political communication. In the
context of Kenya, the adoption of social media for election campaigns
is relatively new compared to their usage in Western democracies.
Most electoral campaigns in the West have deployed social media
since the turn of the millennium. Particularly, the 2004 presidential
campaign in the US demonstrated that the internet can have dramatic
eects on some candidates’ ability to raise campaign resources and
organise activists. Candidates such as Howard Dean and John Kerry
raised tens of millions of dollars and recruited hundreds of thousands
of volunteers through social media campaigns. Both the 2008 and 2012
Obama presidential campaigns demonstrated that social networks
1. Freedom House, “Freedom on the Net 2016: Kenya Country Prole.” Accessed
January 10, 2017. https://freedomhouse.org/report/freedom-net/2016/kenya
2. Kenya ICT Board, “Digital Kenya: A Study to Understand the On-line Life of
Kenyans.” Accessed December 3, 2015. https://www.slideshare.net/tandaaKENYA/
study-of-the-online-life-of-kenyans-2010
3. Kenya’s internet penetration is the highest in Africa. In 2016 Kenya had over 31
million internet users accounting for a 68.4% penetration rate. See Internet World
Stats, “Africa.” Accessed January 6, 2017. http://www.internetworldstats.com/africa.
htm#ke
could be powerful tools for mobilising support.4 In Kenya, however,
social media use for political campaigns began more recently with the
2007 elections, and it was not until the 2013 elections that political
candidates signicantly used social media for political campaigns.5
e central argument of the book is that social media have come to
play an increasingly important role in politics in Kenya in the post-co-
lonial era. My interest lies in studying how Kenyan political candidates
adopt and utilize social media. I seek to understand how social media
adoption inuences the election campaign process. Some writers ex-
ploring the subject of election campaigns in Kenya have indicated the
deployment of a wide range of technologies, some tied to the internet,
for political campaign communications over the years. For instance,
a volume edited by Hyden, Leslie and Ogundimu published in 2002
broadly examines the role of communications in political transition in
Africa by investigating four separate domains: political, technological,
economic, and cultural.6 Research conducted by Makinen and Kuira
in 2008 notes that social media played a remarkable role during the
2008 elections in Kenya, enabling citizen interaction and content
sharing.7 Christa Odinga conducted a study on the role of social media
during the 2007/2008 post-election violence and the general elections
of 2013 and noted that social media deployment increased political
participation and dialogue among Kenyan citizens8. Mukhongo and
Macharia’s 2016 book, Political Inuence of the Media in Developing
4. Shanto Iyengar, Media Politics: A Citizens’s Guide. 3rd ed. (New York: W. W.
Norton, 2016), 137–143.
5. Christa Odinga, “Use of New Media During the Kenya Elections.” Accessed
March 16, 2015. http://www.diva-portal.org/smash/get/diva2:633138/FULLTEXT01.pdf
6. Goran Hyden and Michael Leslie, “Communications and Democratization in
Africa,” In Media and Democracy in Africa, Goran Hyden, Michael Leslie and Folu F.
Ogundimu, eds. (New Jersey: Transaction Publishers, 2002), 8–9.
7. Maarit Makinen and Wangu Kuira, “Social Media and Postelection Crisis in
Kenya,” e International Journal of Press/Politics 13:3 (2008): 328–335.
8. Odinga, “Use of New Media During the Kenya Elections.”
xiv xv
SOCIAL MEDIA & POLITICAL CAMPAIGNS IN KENYA | John O. Ndavula Introduction
Countries, considers the use of media, in general, in political forums in
developing nations.9
While some of these publications have focused on social media pres-
ence and its potential for citizen engagement, few have moved beyond
discussing this prevalence to unpack what truly is at stake with the
emergence of social media use for political campaigns in Kenya. In
other words, while much is known about the existence of social media
as a platform for political communication campaigns, little is known
about how social media are reconstituting and reshaping the traditional
forms of political communication in Kenya. Other important ques-
tions that need to be deeply probed include: (i) Is the Kenyan political
campaign being “transformed” by its contact with social media? If yes,
in what ways? (ii) What are the consequences of social media use for
election campaigns? (iii) What are the theoretical underpinnings of
social media adoption for the political campaign? (iv) What power
do social media exercise in the political campaign process? (v) What
are the implications of social media use for democracy? (vi) What are
the early lessons to be learnt and prospects for the adoption of social
media for election campaigns?
In this book, I use both quantitative and qualitative methods to
succinctly engage with the questions raised above. Using the two
approaches together enabled me to corroborate results, as well as to
elaborate and clarify ndings. More specically, the quantitative ap-
proach gave me a general understanding of political candidates’ social
media adoption and use. It helped illuminate the trends and relation-
ships between social media use and election campaigns. For this quan-
titative phase, I engaged with presidential, gubernatorial, senatorial,
women representative, and parliamentary candidates. I drew a sample
of 338 candidates across the electoral oces, from a population of 2807
candidates who participated in the 2013 general elections. With the
help of research assistants, I administered a standardized question-
9. Lynete L. Mukhongo and Juliet W. Macharia, Political Inuence of the Media in
Developing Countries (USA: IGA, 2016), 1–11.
naire to the candidates. Where the candidates were not available, their
personal assistants responded to the data collection tool. I used key in-
formants to gain access to candidates. I sent some questionnaires over
email to optimize the completion rates. I analysed the quantitative data
obtained from the administration of questionnaires using descriptive
statistics and inferential statistics, more specically the chi-square and
regression analysis, to determine the trends and relationships between
social media use and election campaigns.
e qualitative approach helped me to rene and explain the sta-
tistical results thus obtained. I purposively selected seventeen key
informants, including candidates across the political oces and their
social media site managers, and conducted interviews which helped
to explore in more depth their perspectives on social media adoption
and use for election campaigns. I carried out interviews with selected
political candidates between August 2014 and December 2014 and
conducted content analysis. is analysis helped me to further inter-
pret and elaborate on the quantitative results.
e ndings I obtained helped me to map out how social media are
transforming election campaigns in Kenya. Indeed, at the heart of this
book is a systematic inquiry into the determinants of social media use
for political campaigns in Kenya. e book oers new insight into the
current patterns of social media use. It sheds light on the emerging
patterns of political communication, which in turn may inuence the
practice of political campaigns going forward.
In chapter 1, I chart the history of party political communication
in Kenya. I map the key changes and explain periods that can be
categorized as: pre-modern, modern and post-modern. I demonstrate
the centrality of communication to the election campaign and trace
the emergence and evolution of campaigns’ online presences. I ar-
gue that making sense of the relationship between mass media and
politics requires us to think about the latter’s ever-changing character
and cultural peculiarities. I further posit that seeing the relationship
historically and comparatively between media and politics allows us
xvi xvii
SOCIAL MEDIA & POLITICAL CAMPAIGNS IN KENYA | John O. Ndavula Introduction
to understand and use. Observability is the extent to which candidates
can see and measure the results of social media for election campaigns.
Compatibility relates to how social media are consistent with the past
experiences and needs of the candidates. I move on to discuss envi-
ronmental characteristics that impact a candidate’s decision to adopt
social media. e environment in this context is the social system into
which an innovation is being introduced. e characteristics of the
environment under which candidates adopt social media include level
of oce, competitiveness of the race, constituency type, voter demo-
graphics, and party aliation. I conclude my exploration of the factors
that explain the adoption of social media for election campaigns by
focusing on the technology adoption model. is model tests the in-
terrelationships among adopter characteristics, technological factors,
and environmental factors that lead to the adoption of social media for
election campaigns in the context of Kenya.
In chapter 4 I probe deeper into candidates’ online behaviour by
examining various techniques of social media production for election
campaigns in Kenya. I assess how the four production techniques –
informing, involving, connecting and mobilizing – were utilized in the
2013 election campaigns in Kenya. I examine how the candidates used
the strategies to inform prospective voters, involve supporters, connect
internet users with other political actors, and mobilize advocates in
diering ways and to varying degrees. Specically, I critique how these
techniques impinge on the full utilization of social media for election
campaign communications.
Chapter 5 examines the political impact of social media and as-
sesses what the stakes are for Kenya. I explore the eectiveness of
social media as channels for electoral campaigns in Kenya. I consider
the prospects and challenges of using social media to reach a younger
age demographic. I also situate social media in the larger discourse of
electronic democracy. I re-examine what powers social media can and
do exercise and what the implications are for democracy. I explore the
views of optimists who see social media as progressive and improving
to identify the forces that shape its current form, to realize that it has
not always been like this and that it is unlikely to remain this way.
In this way, the chapter serves as a preview for the chapters to follow
which center on the adoption of social media for political campaigns
in Kenya.
Social media and election campaigns in Kenya are the focus of chapter
2. I rst dene social media as web-based services that allow individu-
als to construct a public prole within a bounded system, articulate a
list of other users with whom they share a connection, and view and
traverse their list of connections and those made by others within the
system. e social media networks I consider in-depth are Facebook
and Twitter. I focus on the intersection between social media and elec-
tion campaigns. I begin by highlighting the general characteristics of
social media, then focus in on specic characteristics of Facebook and
Twitter. is is followed by a discussion on the political implications of
these new modes of communication.
In chapter 3 I interrogate how political candidates come to adopt
social media for election campaigns. I give a theoretical perspective
on the adoption of social media by reviewing diusion of innovations
theory. Diusion of innovations theory explains how society adopts
technologies that are considered new. I consider how the theory can
explain the adoption of social media among candidates. I also consider
the adoption of social media in the 2013 elections at the presidential,
gubernatorial, senatorial, women representative, and parliamentary
levels. I explore the characteristics of the candidates and how they in-
uence adoption of social media. e characteristics explored include
age, gender, education level, awareness of social media, and years of
experience with social media. I then examine certain characteristics
of the technology relevant to its adoption. ese include relative
advantage, complexity, observability, and compatibility. Relative ad-
vantage is the degree to which social media are perceived as superior
in factors such as ease of use, aordability, control, and convenience.
Complexity is the degree to which social media are relatively dicult
Introduction
1
xviii
SOCIAL MEDIA & POLITICAL CAMPAIGNS IN KENYA | John O. Ndavula
the quality of election campaigns, as well as of sceptics who argue that
social media have had little impact on political outcomes. Finally, I
consider whether the payos of social media campaigning are sucient
to advocate for their utilization in election campaign communication
in Kenya.
I conclude in chapter 6 by returning to my original question regarding
the relationship between social media and politics. I use close analysis
of the social media campaign practices and techniques over the course
of the 2013 elections to draw conclusions about the changes in social
media campaigning itself and in campaign communications because of
social media adoption. I consider various potential futures for online
campaigning, taking into consideration the trajectory emerging from
the initial years of practice and the possible developments in political
and technological circumstances. Finally, I focus on the social media
campaign, its implications and trajectory for the future.
In summary, Social Media Adoption and Campaigns in Kenya con-
nects and reects on the many issues and arguments that are raised by
the relationship between social media and politics. I will be exploring
various aspects of that relationship: how the political campaign is af-
fected by the emergence of social media, how to understand factors
impacting technology adoption, how to explain the adoption of social
media by political campaigns in a maturing democracy, and how to
assess the implications and impact of social media on the political
campaign. Without understanding such things, we cannot hope to
comprehend the nature of modern media and modern politics, or to
ascertain whether politics is indeed being transformed by its contact
with these emerging forms of communication.
For the convenience of our readers, most of the data tables have been
printed at the end of each chapter, where readers who want to consult
the tables can do so.
CHAPTER 1
Communication and Political Campaigns
All political parties seek to compete in elections to win and hold
public oce.1 To achieve this, parties seek communication tools that
allow them to reach the masses. Castells argues that politics is based on
a socialised communication, and on the capacity to inuence people’s
minds.2 e main channel of communication between the political
system and the citizens in the immediate past and present has been
the mass media system. In contemporary society, politics is primarily
media politics. e workings of the political system are staged for the
media to obtain the support, or at least the lesser hostility, of citizens,
who then become consumers in the political market.
Perhaps the most obvious transformation in the structure of political
campaigns over time is around technology. Although the additions of
radio in the 1920s and television in the 1950s brought with them sev-
eral alterations to political campaigns, as technological advancements
they were only the beginning. Today, campaigns from the county to
1. Catherine Cook, “Mobile Marketing and Political Activities,” International
Journal of Mobile Marketing 5:1 (2010): 9.
2. Manuel Castells, “Communication, Power and Counter-power in the Network
Society” International Journal of Communication 1:1 (2007): 240.
2
SOCIAL MEDIA & POLITICAL CAMPAIGNS IN KENYA | John O. Ndavula
3
Chapter 1 Communication and Political Campaigns
the national level rely on a range of media strategies, some tied to
recent advances in information and communication technology. In
response to this shi, the nature and structure of campaigns, as well as
the makeup of the people who run them, have changed.3
Chapter 1 charts the history of political communication in Kenya.
It seeks to map the key changes and explain periods that can be cat-
egorized as: pre-modern, modern and post-modern. In this way, the
chapter demonstrates the centrality of communication to the election
campaign and serves as a preview for the chapters to follow. ese ad-
dress social media and election campaigns, diusion of social media in
2013 election campaigns, techniques for campaigning on social media,
and the political impact of social media campaign.
Political Campaigns
e political election campaign is an essential element of a demo-
cratic system. Elections are important because they allow citizens the
freedom to actively participate in selecting their leaders. Not only do
elections give opportunities for participation in determining who will
govern, but they also provide the legitimacy with which to govern.4
e winners of elections receive a general acceptance of their right to
power. Ideally any election can give the winner power, but only a de-
monstrably democratic election will provide the legitimacy necessary
to govern.
In Kenya, elections currently allow citizens to determine occupants
of seats occasioned by the Constitution of Kenya promulgated in
2010. e mainstay of the new constitution is a devolved system of
governance, which necessitated the creation of additional political of-
ces. Kenyans therefore elect candidates not only for the pre-existing
presidential and parliamentary seats, but also for the new gubernato-
3. Judith S. Trent, Robert V. Friedenberg and Robert E. Denton, Political Campaign
Communication: Principles and Practices, 7th ed. (New York: Rowman & Littleeld,
2011), 13.
4. Ibid., 4.
rial, senatorial, women representative, and county seats. Candidates
for these new oces went to the polls for the rst time in the 2013
general elections.
Multi-party democratic elections have been held in Kenya only since
1991. Before then, Kenya was a single-party state. Under Kenya’s rst
post-independence president, Jomo Kenyatta, the ruling KANU party
had tolerated some internal criticism and debate over its platform,
albeit to a gradually diminishing degree. e rst elections, held in
1969, and the subsequent elections in 1974 and 1979 saw the campaign
utilize tactics such as promises of development, trumpeting of achieve-
ments, and leveraging ethnicity, personal abilities and connections.
Clannism or localism, which involved seeking support from specic
clans or geographical parts of a seat, was a key electoral weapon in
most ex-reserves, while ethnic-based campaigning dominated the
settlement zones and the cities.5 A growing criticism of the regime
saw some candidates prevented from campaigning. e situation was
worse in Nyanza province, which was the home of the opposition can-
didate, Oginga Odinga. For instance, the 1969 elections saw the arrest
of Oginga Odinga and a total ban in campaigning in Nyanza province.6
e situation gradually morphed into one in which KANU was de
facto Kenya’s only functioning political party. In 1982, in response to
a failed military coup, Kenyatta’s successor Daniel Moi made a signi-
cant change in the character of Kenyan political life. By introducing
section 2A of the Kenyan constitution, which stipulated that all elected
representatives must be members of the KANU party, Moi transitioned
the country from a de facto to a de jure single-party system. Moi was at-
tempting to stem dissent in KANU by this constitutional amendment.
However, Kenyan citizens protested the change, and this increasing
pressure from the opposition ultimately precipitated the government’s
announcement in December 1991 of the repeal of section 2A. e gov-
5. Charles Hornsby, Kenya: A History Since Independence. (London: I. B. Tauris,
2012), 340.
6. Ibid., 213.
4
SOCIAL MEDIA & POLITICAL CAMPAIGNS IN KENYA | John O. Ndavula
5
Chapter 1 Communication and Political Campaigns
ernment’s agreement to begin registering opposition parties marked
a decisive shi into a new era of multi-party politics in Kenya. Since
then, citizens have been actively engaged in the choice of leadership.
Currently, general elections are held every ve years in Kenya. e
Constitution of Kenya promulgated in 2010 allows voters to choose
candidates for the presidential, gubernatorial, senatorial, women
representative, member of parliament, and county seats. e mainstay
of the new constitution was a devolved system of governance, which
opened the political space by craing more levels of representation.
Kenyans voted under the new law for the rst time in the 2013 general
elections.
Although the campaign period is expected to last for three months
under law, it is common for candidates to begin re-positioning them-
selves long before this. Candidates, therefore, must contend with
long, expensive campaigns before the general elections. During this
extended campaign period, candidates must work to ensure that their
messages are received by the citizens. It is therefore important for us to
examine the current and historical role of communication in election
campaigns.
Communication and Political Campaigns
ere are numerous combinations of economic, sociological, psy-
chological, and historical features that are intrinsic to or reective of
the electoral process. However, the essential core of each campaign
is communication. An electoral campaign can be understood as the
intersecting communication vectors of the candidates, the media, and
the public. Other campaign factors become important in the electoral
system principally through their relationship to the conduits of com-
munication. Communication occupies the area between the goals or
aspirations of the candidate and the behaviour of the electorate, just
as it serves as the bridge between the dreams or hopes of the voter
and the actions of the candidate. Communicating with voters in the
hope of inuencing their behaviour is as old as competitive politics
itself. It follows that, as elections become more competitive, candidates
will continuously seek more ecient means of communicating their
messages.
Political campaigns essentially begin through communication. In-
dividuals verbally announce their intention to run, and posters and
billboards nonverbally announce that election time has begun. During
the campaign, candidates and members of their sta engage in multiple
communicative acts: they debate; appear on television; answer call-in
questions on radio and television talk shows; prepare and present mes-
sages for media commercials; take part in rallies; wear hats; submit
to media interviews; produce web content; and speak at all forms of
public gatherings, from large-scale rallies to intimate neighbourhood
meetings. In addition, they buy radio and television time to further
disseminate their image and message. All this eort is for the single
purpose of communicating with the electorate.7 Essentially then, po-
litical election campaigns are campaigns of communication.
As recent Kenyan elections have shown, close elections oen can
hinge on the persuasion of a relatively small number of voters. is has
brought a recognition of the importance of studying the campaign—
and the role of the mass media in inuencing voters—into sharp focus
in modern politics.
While the value and purpose of elections has remained relatively
constant, the way they are conducted has changed enormously in re-
cent years. In the sections that follow, we must examine those changes
that, to the greatest extent, comprise the essence of the new political
landscape.
Changes in the Political Campaign
e way elections are conducted has changed enormously in recent
years, due in large part to technological advances. It can be argued that
these advances have caused political campaigns to undergo a radical
7. Trent, Friedenberg and Denton, Political Campaign, 16–17.
6
SOCIAL MEDIA & POLITICAL CAMPAIGNS IN KENYA | John O. Ndavula
7
Chapter 1 Communication and Political Campaigns
transformation.8 Bourgault argues that there has been a shi in demo-
cratic thought in Africa, ushered in by the liberalization trend that
began in the early 1990s, and coinciding with the development and
spread of new information technologies, notably the use of computers
and the internet for communication purposes.9 A number of African
countries, among them Kenya, embraced multi-party politics. Indeed,
between 1989 and 2007, more than 140 legislative and 120 presidential
elections were held in forty-three African countries, the overwhelming
majority of which were emerging from decades of single-party regimes
and dictatorships.10 e democratic practice in Africa is moving to-
wards laying emphasis on non-hierarchical, dialogical communication
through popular participation. e implication of this new approach
is that it places communications in a fresh and more central place than
before. In the past, the government went out of its way to control the
ow of news to make information more attuned to what it conceived
as its national priorities, but the current interactive approach facilitates
a discursive process.11
e major changes in political campaign communication can be
viewed as a typology consisting of three stages: the pre-modern, the
modern and the post-modern stage.12 e essential catalyst for the
change is how campaign messages are communicated.
8. Ibid., 4.
9. Louise Bourgault, Mass Media in Sub-Saharan Africa (Bloomington: Indiana
University Press, 1995), 206–225.
10. Marie-Soleil Frère, Elections and the Media in Post-Conict Africa: Votes and
Voices for Peace? (London: Zed Books, 2011), 1.
11. Hyden and Leslie, “Communications and Democratization in Africa,” 8–9.
12. David Farrell and Paul Webb, “Political Parties as Campaign Organizations,”
In Parties without Partisans: Political Change in Advanced Industrial Democracies,
Russell Dalton and Martin Wattenberg, eds. (Oxford: Oxford University Press, 2000),
103–125.
THE PRE-MODERN PERIOD
e pre-modern campaign period spanned the late nineteenth and
early twentieth centuries. e pre-modern campaigns were local, ad
hoc, and interpersonal. ey were characterized by retail politics,
where communication was largely undertaken by willing volunteers
seeking direct contact with voters in a bid to popularize their candi-
dates. Political advertising was notably underutilised.13 e partisan
press was the primary intermediary between political parties and
citizens, and the electorate was characterised by stable social and
partisan alignments.14 ere was limited central coordination since
overarching communication strategies and centralised campaign man-
agement were absent. ere was a relative absence of communication
professionals since political candidates were condent in their own
abilities to understand the electoral “mood.” Consequently, marketing
and public relations also played limited parts in the communication
process. e pre-modern campaign period was also characterized by
non-mediated communication where little thought was given to the
needs of the media which, where permitted, was expected to report
candidates’ speeches more or less verbatim. In addition, campaigns
took place in a narrow timeframe. ere was no sense that campaign-
ing should be extended over a longer period, let alone that it should
become a permanent feature of day-to-day politics.15
e pre-modern period in Kenya coincided with political organiz-
ing to ght for independence from the British colonialists. e colonial
government barred political participation by indigenous groups and
imposed various economic and political restrictions. Many local po-
litical and social organizations emerged to contest the inequity of these
restrictions. Among these early groups were the Kenya African Union
13. Steven Foster, Political Communication, (Edinburgh: Edinburgh University
Press, 2010), 4.
14. Pippa Norris, A Virtuous Circle: Political Communication in Post-industrial
Societies, (New York: Cambridge University Press, 2001), 2–6.
15. Foster, Political Communication, 5.
8
SOCIAL MEDIA & POLITICAL CAMPAIGNS IN KENYA | John O. Ndavula
9
Chapter 1 Communication and Political Campaigns
(KAU) and the East African Association (EAA), the latter of which had
a tenure of only 4 years, from 1921–1925, at which point it was forced
by the colonial government to shut down.16 During these early days,
communication was largely undertaken by willing volunteers seeking
direct contact with citizens. e colonialists controlled the media and
there was no chance for local nationalist or indigenous organizations
to utilize them.
In the rst years aer Kenya’s independence in 1963, political parties
developed an inclination towards one-party governance status. Despite
this, the country embraced and implemented relatively reasonable
democratic practices. Candidates held political rallies and distributed
pamphlets. A handout culture developed, whereby voters expected
to be paid in one form or other to support a candidate. Wealthier
candidates gave the electorate gis of sugar, beer, and cash. ere was
widespread use of harambee (fundraiser) meetings for campaigning.17
Candidates from the ruling party KANU oen combined their party
campaign with normal ministerial activities and tours.18 Some print
publications critical of the government were either banned or denied
revenue from government advertising.19 Although it was vehemently
denied, oathing of the electorate in Central Province was reported in
1969 election campaigns.20
Starting from 1970, however, the parties lost all pretensions to de-
mocracy. Kenya became a one-party state, rst under President Jomo
Kenyatta and then under his successor, Daniel Moi. Political competi-
16. Karuti Kanyinga, “Contestation over Political Space: e State and the Demo-
bilisation of Opposition Politics in Kenya,” in e Politics of Opposition in Contem-
porary Africa, ed. Adebayo. O. Olukoshi (Uppsala: Nordiska Afrikainstitutet, 1998),
39–90.
17. Hornsby, Kenya: A History Since Independence, 276.
18. Walter O. Oyugi, Peter Wanyande and C. Odhiambo-Mbai, e Politics of
Transition in Kenya: From KANU to NARC. (Nairobi: Heinrich Boll Foundation,
2003), 128.
19. Hornsby, Kenya: A History Since Independence, 340.
20. Ibid., 212.
tion was almost entirely suocated and only the Kenya African National
Union (KANU) remained as the dominant party. Later KANU became
the only legal political party aer it engineered constitutional changes
in 1982 to make Kenya a de jure single party state. KANU leaders Jomo
Kenyatta and Daniel Moi used their extensive presidential powers and
control of the media to counter any challenge to their leadership by
the opposition. During the era of the single-party system, the KANU
Government continued and intensied its autocratic rule, political
organization became dicult or altogether impossible, and inter-party
political competition was completely diminished.21 e partisan press
acted in favour of the ruling party’s chosen candidates during elections.
THE MODERN PERIOD
One of the main features of the modern campaign period is the
emergence of distinctly national campaigning. It is characterised by
campaign activities being increasingly coordinated at the central party
level with the help of professional consultants. Television has taken
centre stage as the primary campaign medium and the electorate has
become somewhat more detached from its traditional social and parti-
san ties.22 In the modern period, parties increasingly design their com-
munications around the needs of the print and especially the broadcast
media. Rather than focusing their eorts on communicating directly
with voters, they instead prefer to channel their messages via journal-
ists, compensating for the loss of editorial control with better and more
ecient access to the voters. Political communication in this period
has a strong strategic element which is devised by national campaign
teams and is built around the personalities and characteristics of indi-
vidual party leaders. ere is also a move towards professionalization
of the campaign, as large teams of professional advisors are essential
for modern communication. eir presence is felt throughout the
21. Friedrich E. Stiung, Institutionalizing Political Parties in Kenya (Nairobi:
Friedrich Ebert Stiung, 2010), 6
22. Foster, Political Communication, 12.
10
SOCIAL MEDIA & POLITICAL CAMPAIGNS IN KENYA | John O. Ndavula
11
Chapter 1 Communication and Political Campaigns
campaign, but especially in the areas of political advertising and media
management.
e long campaign is also a characteristic feature of this modern
period of campaign communication. Communication strategies com-
mence signicantly in advance of the ocial campaign. Equipped with
a team of professional advisors, campaign strategies seek to exploit
modern marketing techniques, especially opinion research. National
advertising is also prominent and carefully designed to reinforce the
central campaign messages. Adverts are unlikely to advocate specic
policies but are instead heavily image based and shaped around simple
and emotionally evocative slogans. e goal of press and publicity
ocers (“spin-doctors”) is to set the media’s agenda, ensuring that
the headlines are dominated by stories which play to their party’s
advantage.23 In essence, campaigns become machines for cranking out
mediated communications highly attuned to the needs of the media,
especially television.
In Kenya, the emergence of multi-party politics from 1991 onwards
saw the usage of election campaign strategies characteristic of the
modern period. In 1991, section 2A of the constitution, which had
required all candidates seeking oce to be members of KANU, was re-
pealed and Kenya became a multi-party state. Numerous parties were
formed with a view towards dislodging the KANU party, which had
had a stranglehold on power ever since its ascendance aer Kenya’s in-
dependence. Some of the parties included FORD-Asili, FORD-Kenya
and SAFINA.24 However, the emergence of so many opposition parties
ultimately worked in KANU’s favor, splintering the opposition and
allowing KANU to win with just one-third of the total vote.
In subsequent elections, rather than developing slowly over time
through grassroots participation, parties were formed close to elec-
tions with the sole purpose of capturing and retaining political power.
Although the Constitution of Kenya (2010) acted to institutionalize
23. Ibid., 6.
24. Stiung, Institutionalizing Political Parties in Kenya, 7.
political parties, through basic requirements stipulated by Article 91
and the Political Parties Act (2010), party membership in terms of
ocial registration of citizens to specic political parties continues to
be very low. A national study conducted in 2015 concluded that only
33 percent of respondents belonged to a party, with only 12 percent
paying party membership subscription fees.25 Further, relatively few of
those members who remain consider themselves to be “active.”
e situation of reinventing political parties anew with each elec-
tion cycle leads to an increased reliance on the media to popularize
these newly minted parties to the electorate. In this context, political
communication becomes urgent and intensive. e Kenyan political
communication scene is characterized by the prominence of marketing
professionals, private opinion research, and national advertising. ey
use modern tools of communication such as mass media, radio, TV and
newspapers to advance their messaging. Parties in Kenya use polling
to determine which combinations and associations would work well to
energize their electorate and rely on elaborate spin-doctors to produce
results. Even then, it can be argued that the parties primarily concern
themselves with consolidation and retention of power, rather than on
developing and implementing policies once in oce or undertaking
the other traditional service and patronage roles of political parties.26
FACTORS CONTRIBUTING TO TRANSITION FROM THE
PRE-MODERN TO THE MODERN PERIOD
e transition from the pre-modern period to the modern period
in Kenya occurred because of a combination of factors. Some factors
undermined the viability of pre-modern communications, while oth-
ers reinforced this shi by giving campaign managers positive reasons
for change.
25. Jesuit Hakimani Centre, Fencing Duty Boundaries: Whose Values Matter in
Kenya’s Devolved System? (Nairobi: Jesuit Hakimani Centre, 2015), 17–19.
26. Stiung, Institutionalizing Political Parties in Kenya, 1–2.
12
SOCIAL MEDIA & POLITICAL CAMPAIGNS IN KENYA | John O. Ndavula
13
Chapter 1 Communication and Political Campaigns
First, the rise of multi-party politics in 1991 was a decisive push
factor because it led to the emergence of a de-aligned electorate. is
opened the possibility that voters would be more likely to be open to
persuasion, more indecisive about which party to vote for, and more
likely to switch parties. Under the single-party state, there had been
widespread disengagement from voters, who inevitably met party
propaganda was invariably met with cynical disbelief. Communicating
with such a heavily disinvested electorate presented emerging parties
with serious challenges on a scale not seen before, creating a sharp
need for modern campaign tools.
In addition, the repeal of section 2A of the constitution enabled the
media to be pluralized, with the result that dierent media aligned
themselves with dierent parties. Prior to 1991, there seemed little
incentive in trying to persuade wide sections of a largely stable elector-
ate to change their views. ereaer, in an age when so many factions/
parties emerged, parties had every reason to seek out new groups of
supporters.
e eects of de-alignment were compounded by the new constitu-
tion promulgated in 2010, which reorganized political oces. It brought
in the need to launch campaigns at multiple devolved levels: presiden-
tial, gubernatorial, senatorial, parliamentary, women representative,
and county assembly. e new entrants to the political scene needed
to make themselves known, and hence political communication eorts
were intensied. For instance, political advertising, which came to the
fore in the 2007 election campaigns, intensied signicantly with the
rst elections under a devolved system in 2013.
It is known that tight statutory controls on political campaign
expenditure can seriously hamper the development of expansive com-
munications strategies. However, the absence of laws governing po-
litical campaign communications in Kenya, at least in the early period
of multi-partyism, meant that political parties were free to launch as
aggressive a campaign as their funds allowed. is acted as a pull factor
towards newer modes of communication. Party headquarters were free
to spend what they liked, both before and during election campaigns,
providing of course they could rst raise the money. Major parties
such as the Party of National Unity (PNU) and the Orange Democratic
Movement (ODM) held breakfasts as a way of raising resources. One
major impediment to examining the impact of these events is that the
total expenditure of the parties has never been adequately tracked or
revealed. However, most of parties only buy into the modern media as
much as their nances, and to a lesser extent, their ideologies allow.
For example, the Jubilee Party was able to run on a digital campaign
in part because, having branded itself as “the digital party,” its ideology
and identity supported that strategy.
Parties have consistently demonstrated the ability to accumulate
considerable campaign “war chests,” in part because the sources of their
campaign nancing remain unknown. For instance, a survey done in
Kenya by the Coalition for Accountable Political Financing (CAPF)
showed that presidential candidates in the 2007 general election spent
close to ksh. 6 billion ($75 million) in campaigns. However, with the
2013 elections requiring that six leaders be chosen, the amount for
campaigns was bound to hit ksh 11.2 billion ($ 130 million).27 In the
2013 elections, the total amount spent was estimated at between ksh
8.5–12.7 billion ($100–150 million).28 With these kinds of budgets,
parties have been able to nance activities in marketing, advertising,
and—above all—public relations, which together have redened the
nature of party political communication. ey have engaged in paid TV
for live reportage. e multi-party era and the freeing of the airwaves
nally permitted television and radio journalists to report elections as
27. Mwaura Kimani and Christine Mungai, “Campaign Finance: Price Tag of
Kenya 2012 Presidential Race to Hit $130 Million.” Accessed July 6, 2014. http://
www.theeastafrican.co.ke/news/Campaign-nance--Price-tag-of-Kenya-2012-pres-
idential-race/-/2558/1320582/-/item/0/-/a9r31ez/-/index.html
28. Wangui Maina, “e Money Factor in Race for Kenya’s Top Job.” Accessed
August 8, 2014. http://www.africareview.com/Special-Reports/e-money-factor-
in-race-for-Kenyas-top-job/-/979182/1708124/-/9v0n5/-/index.html
14
SOCIAL MEDIA & POLITICAL CAMPAIGNS IN KENYA | John O. Ndavula
15
Chapter 1 Communication and Political Campaigns
they were happening. Before 1991, broadcasters were prevented from
contemporaneous coverage, especially of opposition parties’ functions.
New generations of politicians responded by spending small fortunes
on securing the skills of media advisers and public relations “gurus.”
e task facing parties was to provide television producers with the
material they required for their programming while at the same time
working to steer the political agenda in the desired direction. Exposure
on peak time television enables candidates to reach more people than
they could in a lifetime of canvassing, handshaking, or addressing
public meetings.29
By 2016, Parliament voted to place the rst statutory limits on
national campaign expenditures. In 2016, the Independent Electoral
and Boundaries Commission (IEBC) of Kenya issued new campaign
nance rules limiting spending by political parties, party contributors,
and aspirants during the 2017 electioneering period beginning in
February. It capped presidential contenders’ spending at a maximum
of Ksh 5.2 billion ($51.8 million); governors, senators, and women
representatives to Ksh 433 million ($4.3 million); and members of par-
liament to Ksh 33.4 million ($330,000). Spending by contestants to the
county assemblies was capped at Ksh 10.3 million ($101,000). e new
rules have also limited political party expenditures during the elections
to a maximum of Ksh 15 billion, and single-source contributors to po-
litical parties to Ksh 3 billion. is is in line with Article 88 (4) (i) of the
Constitution, the IEBC Act 2011, and the Election Campaign Finance
Act 2013, which mandates the electoral body to regulate the amount
of money that may be spent by or on behalf of a candidate or party in
respect of any election.30 All the same, it cannot be credibly argued
that a “ceiling” on campaign spending will seriously compromise the
tradition of ambitious national campaigning.
29. Foster, Political Communication, 13.
30. Walter Menya, “IEBC Caps Political Aspirants, Party Expenditure in 2017
Polls.” Accessed August 11, 2016. http://www.nation.co.ke/news/IEBC-caps-politi-
cal-aspirants-party-expenditure-in-2017-polls/1056-3340778-sxtd31z/index.html
TOWARDS THE POST-MODERN PERIOD
Over the last two decades, political communication has continued
to evolve rapidly, so much so that it is now suggested that a new, post-
modern campaign period is coming into being.31 e rise of the internet
in the post-modern period has led to what has initially been referred
to as an Americanized style of campaigning, where the internet is used
in innovative ways.32 Campaign messages are communicated through
a far wider range of media, particularly through digital technologies:
text messages, emails, and websites.
First, it suggests that electoral behaviour is likely to become even
more volatile and still harder to predict. Recent trends in the spatial
dimension of general election results, most notably the collapse of
uniform swings, already oer supporting evidence of this.33 is has
increased the role of political consultants in political campaigns. is
is particularly true for Kenya, where politicians create and discard
parties at will, using them only as tools to consolidate or gain power.
is has made alliances harder to predict and therefore the electoral
behaviour/shis even more uncertain.
Second, the news media has become fragmented into several chan-
nels, outlets and levels. e electorate has become even further diused
and de-aligned in their voting choices.34 is lends itself to carefully
targeted and personalised campaigning.
ird, it encourages the belief that new generations of voters—the
so-called “digital natives”—cannot be reached through the traditional
means of modern communication: political rallies, mass political
advertising, and broadcast communications. In Kenya, internet user
penetration among the youth (18–24 years) is 61.6 percent.35 Of these,
31. Foster, Political Communication, 16.
32. Andrea Römmele, “Political Parties, Party Communication and New Informa-
tion and Communication Technologies,” Party Politics, 9:1 (2003):7–20
33. Foster, Political Communication, 18.
34. Ibid., 18–19.
35. ITU, Measuring the Information Society (Geneva: ITU, 2013), 143.
16
SOCIAL MEDIA & POLITICAL CAMPAIGNS IN KENYA | John O. Ndavula
17
Chapter 1 Communication and Political Campaigns
digital natives—that is the youth who have ve years or more of using
the internet—are 18.5 percent.36 e digital natives, therefore, are seen
as a growing class of voters who are unwilling to engage politically
through the conventional pathways: by seeking membership in po-
litical parties, organizing for parties, or attending rallies. Rather, these
new digital voters rely heavily on media, especially media tied to the
internet, for political news and information.
Whilst the overarching communication strategy has retained its
dominance, the management of much campaigning has been de-
centralized to give it greater sensitivity to local realities. Devolution
has led to a revival in constituency campaigning or the decentralized
campaign. is has led to targeting and niche marketing. Parties now
deploy the widest range of targeting and niche marketing strategies to
win key “battleground” seats.
Further professionalization has been witnessed in the post-modern
period. Campaign professionals have been deployed to make targeting
and niche marketing possible. Most parties in Kenya have think tanks
that chart the way forward for them. is is especially so in the era of
coalition building.
Permanent campaigning is also a feature of the post-modern period.
e post-modern party increasingly regards each day as a battle to be
won or lost.37 Campaigns in Kenya have turned out to be long drawn,
especially in the wake of party defections, realignments and alliance
building, long before the general election campaigns ocially begin.
is has led to the situation akin to the permanent campaign witnessed
in mature democracies.
36. ITU, Measuring the Information Society, 151.
37. Foster, Political Communication, 19.
Conclusion
e development of election campaigns in Kenya reects increasing
adoption of campaign styles and techniques from Europe and the US,
among them the use of social media. Indeed, technology has played a
signicant role in election campaigns in Kenya, but the use of both old
and new technology has intensied in the 2007–2013 election cycles.
e emergence of the modern campaign, and possibly the post- mod-
ern campaign, in Kenyan politics has not been an even and consistent
process. Election campaign communication strategies have tended to
evolve in “ts and starts.” Invariably, campaign communication strate-
gies have had to await the emergence (and re-emergence) of leaders
willing to follow expert advice. At the same time, understanding of the
main techniques in election campaigns evolves over time, with new
ideas being introduced, dropped, and rened with each contest. It is
evident that political candidates will have to accept that digital technol-
ogy will only deepen the trends towards decentralized campaigning.
For instance, social media have created unprecedented opportunities
for individuals to determine their own sources of news. For Kenya,
which is a developing country, the cost of political communication
during elections can become unsustainable, necessitating a paradigm
shi in strategies for election campaign communication. is leads us
to assess the adoption of social media for election campaigns in the
next chapter.
19
CHAPTER 2
Social Media and Election Campaigns
e rise of social media sites that enable users to create content,
maintain and build social ties, and engage in discussions on public
issues has generated much enthusiasm for political communication.
Social media have radically changed the communicative terrain upon
which forms of political campaigns are shaped. ey fundamentally
challenge long-held assumptions about media reception, participation,
and political roles. e web’s distributed networked environment, there-
fore, forces us to reconceptualise and redene the very term “political
communication” as one that must now account for the ever-expanding
capacity for information storage and retrieval, multiple entry points of
communication, and expanded sites and modes of self-expression.1 In
essence, social media express shiing possibilities for political com-
munication on distributed networks. is chapter, therefore, focuses
on the intersection between social media and election campaigns. It
begins by highlighting the general characteristics of social media and
then focusing on specic characteristics of Facebook and Twitter. is
1. Greg Elmer, Ganaele Langlois and Fenwick McKelvey, e Permanent Cam-
paign: New Media, New Politics (New York: Peter Lang, 2012), 6.
20
SOCIAL MEDIA & POLITICAL CAMPAIGNS IN KENYA | John O. Ndavula
21
Chapter 2 Social Media and Election Campaigns
is followed by a discussion of the political implications of these new
modes of communication.
Social Media
e umbrella term “social media” refers to a range of web-based
tools and services that allow users to create and share content and
information online. ese tools are “social” in the sense that they are
created in ways that enable users to share and communicate with one
another.2 boyd and Ellison dene social media sites as internet-based
applications that allow users to develop a public prole, create a list
of users whom they have a relation with, and view both their own list
of connections and those of others within the system.3 e objective
of social media is to publish large amounts of user-generated content,
using easy-to-use publishing tools. Social media allow producers of
messages to be the consumers also, and receivers to act as distributors.
ere are multiple sites of (re)production and (re)distribution of mes-
sages. Social media draw upon established communication networks
(e.g. interpersonal networks in neighbourhoods, the workplace, or re-
ligious spheres) and established genres of communication (e.g. existing
oratorical traditions, song genres, and parodic styles). ey blend texts
and graphics that derive from both local and transnational sources.
ey function largely as expressive devices in the formation of group
identity, and community or subcultural solidarity.4 In the context of
political campaigns, content may be created by a diverse range of users,
from ocial party sta to interest groups and citizens.
2. Narnia Bohler-Muller and Charl Merwe, “e Potential of Social Media to
Inuence Socio-Political Change on the African Continent,” Africa Institute of South
Africa, brieng no. 46. (2011), 2.
3. Danah Boyd and Nicole Ellison, “Social Network Sites: Denition, History, And
Scholarship,” Journal of Computer-Mediated Communication 13:1 (2008): 211.
4. Debra Spitulnik, “Alternative Small Media and Communicative Spaces,” In
Media and Democracy in Africa, eds. Goran Hyden, Michael Leslie and Folu F.
Ogundimu (New Jersey: Transaction Publishers, 2002), 181.
Although there are many social media sites, I have chosen for the
purposes of this exploration to focus on two of the more robust and
widespread platforms, Facebook and Twitter. While both are popular,
these two platforms dier from each other in terms of the ways their
users can relate to content: the kinds of comments they can make, the
possibility of exporting content from one platform to another and so
on.5 Let us turn our attention to Facebook and Twitter and assesses
their potential for political communication.
FACEBOOK
Facebook was launched in 2004 as a social networking platform.
Facebook’s interface allows its members to create proles with photos,
lists of preferred objects (books, lms, music, cars, political parties),
and contact information. Users can also join groups and communicate
with friends by means of chat and video functions. Several features
channel and mediate social interaction among users, including the
News Feed, for updates of stories from people and pages, the Wall
for (public) announcements, Poke for attracting attention, and Status
for informing others about a user’s whereabouts or for announcing
changes in their (relational, professional) status. Features such as Peo-
ple You May Know (PYMK) help users nd “friends”—connections
and other contacts they may know from overlapping social networks,
job, or school aliations. Facebook automatically noties users about
other people they may be interested in contacting and adding to their
list, with suggestions based on algorithmically computed relationships.
Tagging names to people in pictures helps identify and trace “friends”
across the network. Connectivity is another feature of Facebook that
aims at sharing user data with third parties using Open Graph and the
Like button. Users may deploy the platform to invest in maximum con-
nectedness: the more connections users make; the more social capital
they accumulate. In many ways, Facebook’s connective functions can
5. Elmer, Langlois and McKelvey, e Permanent Campaign, 7.
22
SOCIAL MEDIA & POLITICAL CAMPAIGNS IN KENYA | John O. Ndavula
23
Chapter 2 Social Media and Election Campaigns
be understood as providing empowering and enriching social experi-
ences.6
Facebook’s unique selling point over the years has been its rapidly
growing user base, not only in terms of sheer numbers but also in
terms of diversity and global reach. User demographics, national and
global distribution, and online behaviour are all key to the platform’s
continued relevance. e masses of users signing up for the service
act as proof of the site’s centrality in organizing people’s social lives.
Once a member, the social push to stay connected is enhanced by the
continuous updates about important events that users receive from
other connections.7
In the past few years, Facebook has become a central component of
political activism and campaigning at both local and global levels. e
site is seen not only as a tool for managing political discussion online,
but more importantly, as the means through which political commu-
nication can be transformed into political action. Political activity on
Facebook cultivates a rst-person-framed politics of friendship to en-
ter the realm of socially networked political action and expression.8 In
essence, then, political actors must submit to, adapt to, and make use of
the communicative limits and specic informational logics intrinsic to
Facebook as a social media platform. Political actors emerge not only
in relation to an issue, but also in relation to the platform that allows
them to come into being.
Politicians in African countries have used Facebook extensively in
election campaigns. President Goodluck Jonathan of Nigeria and his
cabinet, as well as the Rwandan president Paul Kagame, were among
6. Jose van Dijck, e Culture of Connectivity: A Critical History of Social Media
(New York: Oxford, 2013), 47–50.
7. Dijck, e Culture of Connectivity, 50
8. Elmer, Langlois and McKelvey, e Permanent Campaign, 47.
the rst heads of state to interact with people through Facebook.9 In
South Africa, political parties use Facebook in the build up to the 2014
election.10 South African political parties, the Electoral Commission,
and civil society stepped up their use of social media to drive the masses
to the polls.11 Botswana’s 2014 elections were characterized by the use
of Facebook. Political parties and individual candidates alike sought
to exploit the opportunity presented by social media to woo voters to
their fold. Similarly, to expand their reach and amplify their voices,
traditional media joined the Facebook hype, a move that legitimised
communication through social media.12
In Kenya, the use of Facebook began, albeit haltingly, in the 2007
elections. A signicant number of politicians in Kenya employed social
media for political marketing in the subsequent 2013 elections.13 All
the presidential aspirants in the 2013 elections, for instance, set up
social media accounts as a means of reaching out to voters directly.
Facebook was used to perform a range of key functions such as opin-
ion formation, interest mediation and party organisation. e winning
political coalition, Jubilee, was very active in their use of social media.
9. Rimini Makama, “Electronic Campaigns and Social Media Elections, e
Frontier of Smart Politics,’ Accessed June 22, 2017. https://medium.com/organizer-
sandbox/electronic-campaigns-and-social-media-elections-the-frontier-of-smart-
of-politics-47db04155405
10. Lauren Tracey, “Will Social Media Inuence Election Campaigning in South
Africa?” Accessed June 22, 2017, https://issafrica.org/iss-today/will-social-media-
inuence-election-campaigning-in-south-africa
11. Tanja Bosch, “Youth, Facebook and Politics in South Africa,” Journal of African
Media Studies, 5:2 (2013): 119–130.
12. Bontle Masilo and Batlang Seabo, “Facebook: Revolutionising Electoral Cam-
paign in Botswana?” Journal of African Elections, 14:2 (2015):110–129.
13. Odinga, “Use of New Media During the Kenya Elections.”
24
SOCIAL MEDIA & POLITICAL CAMPAIGNS IN KENYA | John O. Ndavula
25
Chapter 2 Social Media and Election Campaigns
In Kenya, there are over 4.5 million users on Facebook.14 is
gure accounts for 79 percent of Kenya’s internet users.15 e Kenya
ICT Board indicates that the growth of Facebook usage in Kenya is
manifesting both in urban or infrastructure-rich areas of the country
as well as in rural or infrastructure-poor areas.16 e rapid uptake of
Facebook in Kenya is similar to that of much of the continent. ere
are over 120 million Facebook users on the African continent.17 is
is exponential growth, given that in 2010 there were 17 million users
while in 2009, only 7 million users were recorded.18 All in all, it seems
important to investigate the impact of an entity that touches on and
connects to such a wide swathe of the population. Let us now turn our
attention to Twitter.
TWITTER
Twitter, one of the largest and most popular social media sites, was
launched in 2006.19 Twitter allows users to maintain a public web-based
asynchronous “conversation” through the use of 140-character mes-
sages sent from mobile phones, mobile internet devices, or through
various websites. Twitter’s aim is for users to respond to the question
“What’s happening?” in 140 characters or less.20 ese messages are
termed “Tweets” and are publicly accessible on the user’s prole page
14. Reuters, “Facebook Rakes in Users in Nigeria and Kenya, Eyes Rest of Af-
rica.” Accessed January 6, 2017. http://www.reuters.com/article/us-facebook-africa-
idUSKCN0RA17L20150910
15. Internet World Stats. “e Internet Usage Statistics for Africa.” Accessed De-
cember 9, 2016. http://www.internetworldstats.com/stats1.htm
16. Kenya ICT Board. “Digital Kenya.”
17. Phoebe Parke, “How Many People Use Social Media in Africa?” Accessed Janu-
ary 6, 2017. http://www.cnn.com/2016/01/13/africa/africa-social-media-consumption/
18. Accounting Diary, “Insider Clue on Cellphone Success in Africa.” Accessed
November 4, 2013. http://www.sowareaccountingadvice.com/2010/04/insider-clue-
on-cellphone-success-in.html
19. Dhiraj Murthy, Twitter: Social Communication in the Twitter Age (Cambridge:
Polity, 2013), 1.
20. Ibid., 2.
on the Twitter website. Tweets provide public awareness of all users on
the medium rather than being restricted to one’s friends, as is the case
with Facebook. e dialogue between Twitter users occurs through the
at-sign (@). For example, a user can send someone a message starting
with @ followed by their name. One perceived advantage of Twitter is
that the time commitment required to post a Tweet is minimal in com-
parison to posting a blog or publishing other material on the internet.
Twitter connects tweets to larger themes, people, and groups,
which span a range of associations from sports, music, entertainment,
pop culture and politics. Specically, tweets can be categorized by a
hashtag. Any word(s) preceded by a hash sign (#) are used in Twitter
to note a subject, event, or association. By including a hashtag in one’s
tweet, it becomes included into a larger “conversation” consisting of
all Tweets with the same hashtag. e structure of communication
via hashtags facilitates impromptu interactions of individuals (oen
strangers) in these conversations. It is for this reason that Twitter has
been considered useful in social movements. Hashtags represent an
aggregation of tagged Tweets, and therefore conversations are created
more organically. Just because people are tweeting under the same
hashtag does not mean they are conversing with each other in the tra-
ditional sense. Rather, the discourse is not structured around directed
communication between identied interactants. It is more of a stream,
which is composed of a polyphony of voices all chiming in. Either ser-
endipitously or by reading through scores of Tweets appearing second
by second, individuals and groups interact with each other aer seeing
relevant Tweets. Tweets can be directed to specic individual(s), even
to strangers, using the at-sign. For example, individuals can and do
tweet @uhurukenyatta, the president of Kenya, allowing a conversation
that references him, regardless of whether the president is participat-
ing in that conversation.21
21. Murthy, Twitter, 2–4.
26
SOCIAL MEDIA & POLITICAL CAMPAIGNS IN KENYA | John O. Ndavula
27
Chapter 2 Social Media and Election Campaigns
Candidates can use Twitter to organise people for politics, advo-
cacy, or community awareness.22 e growing use of Twitter through
creating and re-tweeting messages on computers and mobile devices
can be viewed as a major new pattern of communication which can
be exploited for political communication gain. Twitter now plays a
signicant role in facilitating the dissemination of news, especially
using re-tweets, Twitter’s news propagation feature. Popular Tweets
spread very quickly through cascades. Indeed, candidates employing
social networking sites in their campaigns have a potentially higher
probability of reaching voters, besides those who visit their social net-
working sites and attempt to start a viral “chain reaction.” is reach
corresponds to the ways in which campaigning through traditional
mass media was able to reach passive “viewers.”23
In Africa, Twitter was used for political mobilization that was suc-
cessfully done in Egypt, Tunisia and Libya where revolutions and
coups took place. Activists in these countries used Twitter to plan and
execute the revolutions.24 In the 2007 Nigerian elections, politicians
used Twitter to popularize their candidature. However, the platform
was considered elitist.25 In Ivory Coast Twitter gained prominence in
2010 during the socio-political crisis. A similar trend in Twitter usage
was recorded in South Africa, in the build up to the 2014 election.26
22. Jennifer Evans-Cowley and Justin Hollander. “e New Generation of Public
Participation: Internet-Based Participation Tools.” Planning Practice & Research 25:3
(2010), 397–398.
23. Natch Greyes, “e Untapped Potential of Social Media: A Primer for Savvy
Campaigners,” Campaigns and Elections 300 (2011), 45–47.
24. Nawaf Abdelhay, “e Arab Uprising 2011: New Media in e Hands of a New
Generation in North Africa,” Aslib proceedings: New Information Perspectives, 64:5
(2012): 529–539.
25. Alex Laverty, “ICT, Social Media and Elections in Africa: A Prospective
Study,” Accessed July 22, 2017. https://theafricanle.com/ict/ict-social-media-and-
elections-in-africa-a-prospective-study/
26. Tracey, “Will Social Media Inuence Election?”
In Kenya, Twitter was rst used in the 2007 election campaigns.
However, the platform was not popular with most of the candidates.27
Aerwards, there was a signicant increase in the use of Twitter for
political mobilization. Twitter usage peaked in the 2013 elections in
Kenya. Presidential candidates as well as candidates for other electoral
oces used Twitter for voter mobilization. President Kenyatta was an
active tweeter during the 2013 election campaigns in Kenya.28 In rank
by country, Kenya is placed as the fourth most active on Twitter in
Africa (behind only Egypt, Nigeria, and South Africa) with about 125
million geolocated Tweets.29
COMPARING TWITTER AND FACEBOOK
Twitter is oen compared to Facebook and sometimes considered
a public version of the latter. is comparison has some truth to it.
Both media are social, tend to elicit regular contributions that are not
verbose, and are highly interactive. However, the two media are unique
in many important ways. First, Facebook provides web services which
facilitate users maintaining a public or semi-public prole within a
bounded system and through which they can articulate a list of other
users with whom they share a connection.30 On the other hand, Twit-
ter enables anyone to publish and access information, collaborate on
a common eort, or build relationships.31 In other words, Twitter’s
emphasis is not as “bounded” to communities of friends as Facebook
is. Rather, Twitter is a publishing-oriented medium. ough Facebook
27. Odinga, “Use of New Media During the Kenya Elections.”
28. Frida Orring (2013). Kenyan Elections 2013 – a tech fail or not? Freedom on the
Internet. Accessed July 4, 2014. http://blog.swedenabroad.se/fxinternet/2013/05/10/
kenyan-elections-2013-a-tech-fail-or-not/
29. Portland, “How Africa Tweets.” Accessed January 15, 2017. http://portland-
communications.com/publications/how-africa-tweets-2015/
30. Boyd and Ellison, “Social Network Sites,” 211.
31. Arthur L. Jue, Jackie Alcalde Marr, and Mary Ellen Kassotakis, Social Media
at Work: How Networking Tools Propel Organizational Performance (San Francisco:
Jossey-Bass, 2010), 4.
28
SOCIAL MEDIA & POLITICAL CAMPAIGNS IN KENYA | John O. Ndavula
29
Chapter 2 Social Media and Election Campaigns
does multicast, this is not its emphasis, per se. Twitter, in many ways,
shares more similarities with blogs, albeit the posts on Twitter are
considerably shorter. However, once one’s tweets are aggregated, a new
structure emerges. is is not merely a technical consideration, but
rather a recognition that the organization of communication as a series
of short communiques is qualitatively dierent from examining Tweets
individually. As a corpus, they begin to resemble a more coherent text.
Granted, the corpus is disjointed, but narratives can and do emerge.
For this reason, Twitter is best considered as a microblog, a blog that
consists of short messages rather than long ones.32 It is also important
to note that Facebook and Twitter are not mutually exclusive. In other
words, a candidate could choose to use both sites for an election cam-
paign.
Social Media and Election Campaigns
All over the world, social media are being used for political com-
munication purposes. Social media oer a greater democratic space
for political candidates seeking to engage with citizens.33 Social media
have emerged as a space where citizens can be informed and mobilized
on issues of common interest, come together as a public, and inuence
political decision-making by engaging in a wide range of institutional
and non-institutional practices—from voting to online and oine
protests. is description falls in line with some of the more optimistic
discourse equating social media with democratic participation, and
which revives the notion of the public sphere.34 Indeed, Balkin argues
that social media use has its roots in democratic culture. He further
posits that “a democratic culture is a culture in which individuals have
a fair opportunity to participate in the forms of meaning-making that
constitute them as individuals. Democratic culture is about individual
32. Murthy, Twitter, 8
33. Michel Marriott, “Blacks Turn to Internet Highway, and Digital Divide Starts
to Close” New York Times, March 31, 2006, A1.
34. Elmer, Langlois and McKelvey, e Permanent Campaign, 47–48.
liberty as well as collective self-governance; it is about each individual’s
ability to participate in the production and distribution of culture.”35
In any democratic society, civil liberties such as participatory rights—
freedom of expression, association, and assembly—are essential for the
viability and durability of such a society. Indeed, in Kenya, democratic
rights are enshrined in the Constitution, which provides for, among
other things, the right and freedom of the individual, conscience,
expression, movement, assembly, and association. Social media may
thus provide an interactive avenue for the expression of these core
participatory rights and freedoms, as well as an alternative forum for
the public to engage with candidates during election campaigns.
In Africa, the ongoing eorts to enhance democracy have benetted
from the liberalisation of mass media. Hyden and Leslie argue that
social media have signicant potential to mediate between state and
society in contemporary Africa. Since formal media are still not wholly
free, the internet is signicant in helping people create meaningful
communicative spaces for themselves.36 In Kenya, while political
pressures can still be brought to bear to silence or constrain the main-
stream media, there is unrestricted access to social media sites such
as Facebook and Twitter, and even fewer incidences of censorship on
social media.37 Social media essentially form platforms for alternative
spheres of communication between candidates and voters, which
candidates are increasingly learning to utilise and incorporate into
their larger campaign strategies. Several candidates in Kenya employed
social media for political communication in the 2013 elections. e
following section looks at the usage of social media in the 2013 elec-
tions in Kenya, comparing usage across dierent demographics and
political party aliations.
35. Jack M. Balkin, “Digital Speech and Democratic Culture: A eory of Freedom
of Expression for the Information Society,” New York University Law Review, 79:1
(2004), 4.
36. Hyden and Leslie, “Communications and Democratization in Africa,” 17.
37. Freedom House, “Freedom on the Net 2016.”
30
SOCIAL MEDIA & POLITICAL CAMPAIGNS IN KENYA | John O. Ndavula
31
Chapter 2 Social Media and Election Campaigns
Adoption of Social Media in the 2013
Election Campaigns in Kenya
e use of online media as a means of political communication is
not new in Kenya, having been used in the December 2007 elections.38
Even in the preceding election in 2002, the major political parties and
some individual candidates had an online presence. Previous elections
in Kenya were dominated by rallies and speeches, and spending by
contenders was largely on direct gis of various kinds to the elector-
ate. During the 2013 elections, the growing trend towards adoption of
social media for election campaigns that had begun in the 2007 elec-
tions gained signicant momentum. As an indication of the increased
prominence of social media, most candidates (87.3 percent) used
social media for political communication in the 2013 elections (see
Figure 2.1).
It is worthwhile to note that the adoption of social media increased
from being extremely sporadic in the 2007 elections, to a high user in-
cidence among candidates in the 2013 elections. To achieve this level of
adoption within a ve-year period is quite remarkable for a developing
nation with a maturing democracy. is indeed shows that there has
been a paradigm shi in the ways in which political candidates express
themselves in the political realm ever since the emergence of new me-
dia. e internet is providing political actors with new spaces in which
to articulate a variety of political information, and subsequently social
media marketing is rapidly becoming a signicant communications
channel for reaching the public.
For the rst time, all the presidential aspirants in the 2013 elections
set up social media accounts as a means of reaching out to voters
directly. Social media were used to perform a range of key functions
such as opinion formation, interest mediation, and party organisation.
Some parties and candidates stressed downward dissemination of
information via new media, whilst others emphasised their interactive
38. Odinga, “Use of New Media During the Kenya Elections.”
and targeting possibilities. e winning political coalition, Jubilee, was
very aggressive in the use of social media. Jubilee’s self-nomination
as the “digital party” was not necessarily invented to reect an active
social media presence, but rather hinged on their political manifesto to
boost the Kenyan ICT sector. Nevertheless, they clearly invested heav-
ily in social media. President Uhuru Kenyatta was an active tweeter,
having received mention as one of Africa’s top ten tweeting politicians
by the British newspaper e Guardian.39 He set up an elaborate social
media team with Dennis Itumbi at its helm. e team managed to
inuence discourse and successfully brand Kenyatta’s Jubilee Coalition
as the “digital party.”
Diusion of innovations theory suggests that a technology is more
likely to be adopted if it is perceived as being better than the technol-
ogy it supersedes by a particular group of users, measured in terms that
39. Orring, “Kenyan Elections 2013—A Tech Fail or Not?”
Figure 2.1: Candidates’ Adoption of Social Media in 2013 Elections
32
SOCIAL MEDIA & POLITICAL CAMPAIGNS IN KENYA | John O. Ndavula
33
Chapter 2 Social Media and Election Campaigns
matter to those users, like cost advantage, social prestige, convenience,
or satisfaction.40 erefore, it is possible that the high adoption inci-
dence could be explained by the perceived benets that social media
oered political actors.
Candidates interviewed recognised some clear advantages that social
media hold over traditional campaign tools. ey oen referenced the
importance of social media’s niche. One member of the social media
campaign team for a presidential candidate observed:
“Our campaign strategy was to target the youth who had most
votes. We looked for the easiest way to reach the youth and
found social media a vibrant platform to run on.”
In addition, a senatorial candidate observed that:
“Social media makes it much easier to communicate because
there are voters who may not attend rallies but since most
people can access at least a handset which is connected to the
internet, then denitely the message will easily reach the elec-
torate. ough we cannot underrate personal contact, which
door to door or meeting in small gatherings like churches give,
using social media makes work easier. Using social media also
helps to satisfy a class of techno-savvy voters who access almost
everything else through social media.”
Other campaigns were drawn by the low cost of using social media.
A parliamentary candidate stated:
“e biggest benet of Facebook is that it is free. It is one of the
cheapest ways to connect with voters.”
Other contestants were drawn by the local and international reach of
social media. One gubernatorial candidate observed:
“Social media is [sic] able to reach people living in the diaspora,
who in turn inuence voters back in the home constituency.”
40. Everett M. Rogers, Diusion of Innovations, 5th ed. (New York: Free Press,
2003), 12–18.
Other contestants commented that they adopted social media
because of their synchronous quality. One parliamentary candidate
stated that:
“Social media reaches [sic] the voters in real time. Once you
create a post, it is immediately out there. You also get quick
response from the electorate. You can just get a post or a ques-
tion or a response from a particular person and you follow it
through with that person.”
Another social media strategist for a presidential candidate posited
that:
“It is easier to reach people through social media. ere are
people who receive social media post notications and they are
able to follow what is trending in real time.”
Indeed, political movements powered by social media have toppled
governments as well as helped others to win elections. e unprec-
edented wave of political change seen in the Arab nations, especially
in Tunisia, Egypt, and Libya, has been accredited to social media.41
Dictatorial regimes in Northern Africa were replaced by democratic
ones through the use of social media as platforms for political par-
ticipation.42 Chief among the social media platforms employed for
political mobilization were Facebook and Twitter.
Among the studies which support these ndings are those by Evans-
Cowley and Hollander; Marriot; Römmele; and Tedesco, Miller and
Spiker, which conclude that politicians are increasingly incorporating
41. Abdelhay, “e Arab Uprising 2011,” 529–539.
42. Philip N. Howard and Malcolm R. Park, “Social Media and Political Change:
Capacity, Constraint and Consequence,” Journal of Communication, 62 (2012):
326–359.
34
SOCIAL MEDIA & POLITICAL CAMPAIGNS IN KENYA | John O. Ndavula
35
Chapter 2 Social Media and Election Campaigns
social networking sites in their campaigns.43 Additionally, studies
conducted by Gulati and Williams show that a record number of con-
gressional candidates maintained a campaign website in US elections
in 2006.44 ey further argue that there is increased standardisation
of baseline informational web content and features, as well as greater
integration of websites into candidates’ overall communication strate-
gies.
However, some of the 12.7 percent of candidates who did not use
social media cited constituency characteristics as reasons for not us-
ing social media. Some argued that since their constituencies were
predominantly rural and poor, voters therefore were most unlikely
to use social media. Diusion of innovations literature suggests that
constituency factors account for the dierence in adoption of new
technologies because potential adopters are mindful of the degree to
which an innovation is compatible and incompatible with expectations
(existing norms and values), as well as the needs and capacities of its
users or customers.45 Interviews with political candidates and sta on
43. Evans-Cowley and Hollander, “e New Generation of Public Participation,”
397–398.; Marriott, “Blacks Turn to Internet Highway,” A1; Römmele, “Political Par-
ties,” 7–20.; John C. Tedesco, J. L. Miller and J. A. Spiker, “Presidential Campaigning
on the Information Superhighway: An Exploration of Content and Form” in e
Electronic Perspective on the 1996 Campaign Communication, eds. Lee L. Kaid and
Dianne. G. Bystrom (Mahwah, NJ: Lawrence Erlbaum, 1999), 51–63.
44. Girish. J. Gulati and Christine B. Williams, “Closing Gaps, Moving Hurdles:
Candidate Website Communication in the 2006 Campaigns for Congress,” Social
Science Computer Review 25:4 (2007): 443–465.; Williams, Christine. B., and Girish
J. Gulati, “Social Networks in Political Campaigns: Facebook and the Congressional
Elections of 2006 and 2008.” New Media and Society 15:1 (2012): 61–67.; Girish. J.
Gulati and Christine B. Williams, “Social Media and Campaigns 2012: Developments
and Trends for Facebook Adoption.” Social Science Computer Review 31:5 (2013):
577–588.
45. Louis G. Tornatzky and Katherine J. Klein, “Innovation Characteristics and
Innovation Adoption Implementation: A Meta-Analysis of Findings,” IEEE Transac-
tions on Engineering Management, 29:1 (1982), 28–45.; Stephen Ward and Rachel
Gibson, “European Political Organisations and the Internet: Mobilisation, Participa-
tion, and Change,” in Routledge Handbook of Internet Politics, eds. A. Chadwick and
Philip N. Howard (New York: Routledge, 2009), 25–39.
their social media campaign management teams seemed to support
this view. For example, one parliamentary candidate stated:
“Our campaign did not utilise Facebook or other social net-
working sites as part of our campaign. Our outreach eorts
focused largely on grassroots methods such as door-to-door
campaigning, use of posters and other personal interaction
with voters.”
TYPE OF SOCIAL MEDIA USED IN THE 2013 ELECTION CAMPAIGNS
Facebook and Twitter are the most common type of social media
used in Kenya.46 Findings obtained from the analysis of the 2013 elec-
tion campaigns showed that 78 percent of candidates used Facebook
in their elections campaign, while only 5.6 percent used Twitter (see
Figure 2.2). e ndings indicated that Facebook diused more rapidly
than Twitter. One reason that accounts for this trend is that Facebook
is considered easy to use compared to Twitter. Whereas Facebook is
considered a communication platform for the masses, Twitter is con-
sidered an elite platform. Also, Facebook provides more exibility in
terms of type of messages and size while Twitter is necessarily more
concise.
A possible reason accounting for the lower adoption rates for Twit-
ter is that it requires technical expertise and specialised knowledge to
leverage eectively. In addition, the micro-blogging nature as well as
the technical skills needed to eectively use Twitter may discourage
potential adopters. For instance, Twitter uses a well-dened mark-up
culture and a well-dened mark-up vocabulary which conveniences
users with brevity in expression but makes it challenging for potential
adopters to utilise eectively for election campaigns.
In addition, research indicates that Facebook has more reach than
Twitter in Kenya.47 It is therefore no surprise that a similar trend was
46. Kenya ICT Board, “Digital Kenya.”
47. Kenya ICT Board, “Digital Kenya.”
36
SOCIAL MEDIA & POLITICAL CAMPAIGNS IN KENYA | John O. Ndavula
37
Chapter 2 Social Media and Election Campaigns
witnessed on the political front, where Facebook was the site of choice
for most candidates.
In general, the rapid diusion of social media is due to the fact that in
Kenya there is unrestricted access to social media sites such as Facebook
and Twitter, and fewer incidences of censorship on social media. e
ease with which a candidate can create a social media account shows
that technology has acted to level the campaign eld. Even the poorest
candidate has the capacity to publicize his or her candidacy online.48
Seen in this way, then, social media essentially provide platforms for
alternative spheres of communication between candidates and voters,
allowing new voices to emerge.
TYPE OF SOCIAL MEDIA AND LEVEL OF OFFICE
ere are several levels of political oce designated by the constitu-
tion of Kenya (see chapter 1). However, I chose to focus on the type
48. Iyengar, Media Politics, 144–145.
of social media used at the presidential, gubernatorial, senatorial,
women representative, and parliamentary levels. Data indicate that
all presidential candidates used Facebook (see Figure 2.3). Candidates
at the gubernatorial and parliamentary levels reported a high usage
incidence as well. Senatorial candidates posted a comparatively lower
usage at 67.9 percent. In contrast, Twitter was the least used site. Twit-
ter’s highest usage stood at a paltry 17.9 percent among senatorial
candidates. Chi-square tests conducted indicated that the variation in
the adoption of social media across the political oces was statistically
signicant (X2 = 21.034, df = 8, Sig = 0.007).
e inference that can be made from these ndings is that all the
candidates, irrespective of oce, preferred Facebook to Twitter. Can-
didates identied the social media platform most popular with Kenyan
voters and concentrated their eorts on that medium. In general, the
results indicated that candidates for new oces (gubernatorial, senato-
rial, and women representative) as well as for pre-existing oces (pres-
Figure 2.2: Type of Online Media Used by Candidates
Figure 2.3: Relationship between Electoral Oce and Type of Online Media
38
SOCIAL MEDIA & POLITICAL CAMPAIGNS IN KENYA | John O. Ndavula
39
Chapter 2 Social Media and Election Campaigns
idential and parliamentary) equally exploited available technological
resources to help them expand their electoral base and maximize voter
turnout among their supporters.
Interviews that I conducted indicate that candidates recognised clear
advantages associated with using social media. Candidates referenced
the importance of Facebook’s niche. One parliamentary candidate
stated that:
“We are trying to reach younger voters, and most young people
use Facebook.”
is fragmentation of channels is essentially a practical extension of
post-modern campaign communication.49
Other campaigns were drawn by low cost. One parliamentary can-
didate observed that:
“e biggest benet of Facebook is that it is free. So, in terms
of value, it is probably one of the best things that we can use to
connect to voters.”
Interviews with political candidates, as well as with sta on their
social media campaign teams for the 2013 general elections, seemed to
indicate that Kenya is still a middle-transition state in terms of social
media use. One parliamentary candidate stated that:
“Social media is [sic] the marketing platform for the future.
Most voters in the future will be young people who are well
educated and who have access to the internet. We cannot
separate technology from political marketing. e internet
will play a key role in the future especially with the advent of
integrated marketing communication (IMC).”
Diusion of innovations literature suggests an additional reason for
variation in usage by stating that constituency factors have a bearing
on rates of adoption of new technologies. is implies that political
candidates in Kenya are mindful of the degree to which an innovation
is compatible or incompatible with the needs and capacities of its users,
49. Norris, “A Virtuous Circle,” 2–23.
and hence adopt it with variations. Variegated adoption in relation to
constituency characteristics is a view supported by Tornatzky & Klein,
and Ward & Gibson.50
A parliamentary candidate observed that:
“Social media has [sic] been overrated. We are still very
‘analogue’ so to speak. We still prefer the spoken word, we still
prefer the human contact that rallies provide.”
Social media act to supplement, rather than replace, conventional
media in election campaigns. is would suggest that candidates
who are the most likely to embrace social networking sites are those
who see this new communication medium as an additional tool for
winning votes.51 Additionally, researchers have noted that politicians
have increasingly begun to employ long-tail marketing strategies in
their campaigns.52 e logic of long-tail marketing is to “sell less of
more,” which in a political campaign context entails that candidates
spread their eorts across several dierent channels, each with specic
intended target groups and tailored messages.
50. Tornatzky and Klein, “Innovation Characteristics and Innovation Adoption
Implementation,” 28–42.; Ward and Gibson, “European Political Organisations and
the Internet,” 2–12.
51. Williams and Gulati, “Social Networks in Political Campaigns,” 66–67.
52. Chris Anderson, e Long Tail: Why the Future of Business is Selling Less of
More (New York: Hyperion, 2006), 15–27.; Josh Koster, “Long-Tail Nanotargeting: Al
Franken’s Online Ad Buys Earned an Unbelievable Return on Investment,” Politics,
Feb (2009), 23–26.; Abigail Shaha, “Long Tail Politics: Sell Less of More,” Politics, Nov
(2008), 14–15.
4140
SOCIAL MEDIA & POLITICAL CAMPAIGNS IN KENYA | John O. Ndavula
Conclusion
e outcome of the investigation demonstrates that most political
candidates used social media during their election campaign in the
2013 general elections. is level of adoption for a developing nation
with a maturing democracy indeed shows that there was a paradigm
shi in the ways in which contestants market their candidature in the
context of general elections. e development of political communi-
cation in Kenya reects increasing adoption of campaign styles and
techniques from Europe and the US, and the uptake of social media for
political communication is a key component of this shi.
ese ndings show that new forms of political communication are
becoming part of the Kenyan political culture. However, rallies and
speeches, and spending by contenders largely on direct gis of various
kinds to the electorate, are still very much a part of the modern cam-
paigns in Kenya. Overall, the use of online media as a means of political
communication is gaining ground among political candidates, but it is
not yet a replacement for other traditional forms of campaign com-
munication. Hence, social media serve as an add-on to other campaign
eorts geared towards election campaigns.53 is chapter highlighted
the adoption of social media for election campaigns in Kenya. In the
next chapter, I explain how the diusion of innovations theory can
help us to understand adoption of social media. I highlight the critical
factors contributing to the adoption of social media in the context of
political communication. I will also examine the extent to which social
media were deployed and elaborate on the factors that accounted for
their adoption in the 2013 election campaigns in Kenya.
53. Maria L. Sudulich et al., “Me Too for Web 2.0? Patterns of Online Campaign-
ing among Candidates in the 2010 UK General Elections.” Accessed April 26, 2014.
http://ipp.oii.ox.ac.uk/sites/ipp/les/documents/IPP2010_Sudulich_Wall_Jansen_
Cunningham_Paper.pdf.
CHAPTER 3
Diffusion of Social Media in the 2013
Election Campaigns
e extent to which social media campaigning practice has been
employed across campaigns varies widely. In this chapter I explore
the factors that account for the diusion of social media in the 2013
election campaigns in Kenya. Diusion is dened as the process by
which a technology is adopted and gains acceptance by members of
a certain community. Adoption decisions are thought to depend on
characteristics of the adopter, characteristics of the technology, and
characteristics of the environment.1
I shall rst give a theoretical perspective on the adoption of social
media by reviewing the diusion of innovations theory. Diusion of
innovations theory explains how society adopts technologies that are
considered new.
Second, I shall explore the characteristics of the candidates and how
they inuence adoption of social media. e characteristics I focus on
are age, gender, education level, awareness of social media, and years
of experience with social media.
1. Rogers, Diusion of Innovations, 1–38.
42
SOCIAL MEDIA & POLITICAL CAMPAIGNS IN KENYA | John O. Ndavula
43
Chapter 3 Diffusion of Social Media in the 2013 Election Campaigns
ird, I shall look at characteristics of the technology relevant to its
adoption. ese include relative advantage, complexity, observability
and compatibility. Relative advantage is the degree to which social
media are perceived as better across factors such as ease of use, aord-
ability, control, and convenience. Complexity is the degree to which
social media are relatively dicult to understand and use. Observabil-
ity is the extent to which candidates can see and quantify the results of
social media as they impact election campaigns. Compatibility relates
to how social media are consistent with the past experiences and needs
of the candidates.
Fourth, I shall discuss characteristics of the environment which
impact candidates’ decision to adopt social media. e environment
is the social system into which an innovation is being introduced. I
enumerate and explore the characteristics of the environment under
which candidates adopt social media, including level of oce, com-
petitiveness of the race, constituency type, voter demographics, and
party aliation.
I conclude my exploration of the factors that explain the adoption
of social media for election campaigns by focusing on the technology
adoption model. is model tests the interrelationships among adop-
ter characteristics, technological factors, and environmental factors
that lead to the adoption of social media, specically as they pertain to
election campaigns in the context of Kenya. Let us turn our attention
to the diusion of innovations theory.
Diffusion of Innovations Theory
Diusion of innovations theory was developed to explain how socie-
ties come to adopt technologies. e theory was proposed by Everett
Rogers in 1962, and seeks to explain how technology is taken up by a
specic population.2 Various scholars have used the diusion of inno-
vations theory to explain social media adoption in political campaigns.
2. Rogers, Diusion of Innovations, 5–23.
Foot and Schneider conducted studies in 2006 using the diusion of
innovations theory to explain web campaigning in American elec-
tions.3 In 2010, Peter Chen conducted a study examining the adoption
and use of digital media in election campaigns in Australia, Canada,
and New Zealand. He employed diusion of innovations theory in his
study.4 More recently in 2012 and 2013, Gulati and Williams conducted
various studies on social media and campaigns in America using the
diusion of innovations theory as a frame of reference.5 erefore,
scholars continue to nd the diusion of innovations theory relevant
in examining social media and political campaigns.
Diusion is dened as the process by which a technology is adopted
and gains acceptance by members of a certain community. e stages
by which a person adopts an innovation, and whereby diusion is ac-
complished, include awareness of the need for an innovation, decision
to adopt (or reject) the innovation, initial use of the innovation to test
it, and continued use of the innovation.6
Given that these decisions are not collective, each member of the
social system faces their own innovation decision that follows a ve-
step process (see Figure 3.1).7 According to Rogers, the rst step is
knowledge, where a person becomes aware of an innovation and has
some idea of how it functions. e second step is persuasion, where a
person forms a favourable or unfavourable attitude toward the innova-
tion. e third step is decision, which involves a person engaging in
activities that lead to a choice to adopt or reject the innovation. e
fourth step is implementation; in this stage, the individual uses the
innovation to a varying degree depending on the situation. During this
3. Foot and Schneider, Web Campaigning, 170–175.”
4. Peter Chen, “Adoption and Use of Digital Media in Election Campaigns: Aus-
tralia, Canada and New Zealand.” Public Communication Review, 1 (2010): 3–26.
5. Gulati and Williams. “Social Media and Campaigns 2012,” 577-588.; Williams
and Gulati, “Social networks in political campaigns,” 52–71.
6. Les Robinson, “A summary of Diusion of Innovations.” Accessed December 9,
2016. http://www.enablingchange.com.au/Summary_Diusion_eory.pdf
7. Rogers, Diusion of Innovations, 169–170.
44
SOCIAL MEDIA & POLITICAL CAMPAIGNS IN KENYA | John O. Ndavula
45
Chapter 3 Diffusion of Social Media in the 2013 Election Campaigns
stage, the individual determines the usefulness of the innovation and
may search for further information about it. e nal step is conrma-
tion. In this stage, the individual nalises their decision to continue us-
ing the innovation and may use the innovation to its fullest potential.8
en, diusion of innovations theory proposes ve constructs that
inuence the adoption of any innovation. ese are relative advantage,
compatibility, complexity, trialability, and observability. According to
Rogers, these characteristics determine between 49 and 87 percent of
the variation in the adoption of new technology. ese ve qualities
create a valuable frame of reference for the investigation into the adop-
tion of social media for election campaigns.
Further, adoption decisions are thought to depend on (i) charac-
teristics of the adopter, (ii) characteristics of the technology and (iii)
characteristics of the environment. To begin with, adopter characteris-
tics that inuence technology adoption are nancial security, opinion
8. Ibid., 169–218.
leadership, age, personal condence, level of information, and attitude.
ese characteristics then further dene categories of adopters: inno-
vators, early adopters, early majority, late majority and nally, laggards.
Innovators are the rst to try a technology and are willing to take risks.
Early adopters, who are second, are on the lookout for a strategic leap
forward in their activities and are quick to make connections between
clever innovations and their needs. Early adopters also tend to enjoy
leadership roles. e early majority are cost sensitive and risk averse.
e late majority are conservative pragmatists who hate risks and are
uncomfortable with any new idea. Laggards, who come in last, see a
high risk in adopting a technology.
Second, characteristics of the technology relevant to its adoption
include relative advantage, compatibility, complexity, trialability, and
observability. Rogers denes relative advantage as the degree to which
the technology is perceived as better than the technology it supersedes
by a group of users, measured in terms that matter to those users,
like cost advantage, social prestige, convenience, or satisfaction. e
greater the perceived relative advantage of an innovation, the more
rapid its rate of adoption is likely to be. Compatibility is the degree to
which an innovation is perceived as being consistent with the values,
past experiences, and needs of potential adopters. An idea that is in-
compatible with their values, norms, or practices will not be adopted
as rapidly as an innovation that is compatible. Rogers denes com-
plexity as “the degree to which an innovation is perceived as relatively
dicult to understand and use.”9 In contrast to the other attributes,
complexity is negatively correlated with the rate of adoption. us,
excessive complexity of an innovation is an important obstacle in its
adoption. Trialability is the degree to which an innovation may be
experimented with on a limited basis. An innovation that is trialable
presents less uncertainty to the individual who is considering it. Trial-
ability is positively correlated with the rate of adoption. Rogers denes
observability as “the degree to which the results of an innovation are
9. Ibid., 15.
Figure 3.1: The Process of Diffusion of Innovation
46
SOCIAL MEDIA & POLITICAL CAMPAIGNS IN KENYA | John O. Ndavula
47
Chapter 3 Diffusion of Social Media in the 2013 Election Campaigns
visible to others.”10 e easier it is for individuals to see the results of
an innovation, the more likely they are to adopt it. Visible results lower
uncertainty and stimulate peer discussion of a new idea.
ird, adoption decisions are thought to also depend on character-
istics of the environment. e environment is described as the nature
of the social system into which the innovation is being introduced.
According to Foot and Schneider, the political environment includes
the members of a particular community, their level of income, level
of oce, competitiveness of the race, and party aliation.11 Let us ex-
amine adopter characteristics and how they inuenced social media
adoption in the 2013 general elections in Kenya.
Adopter Characteristics and Social Media Adoption
In this section I examine the characteristics of the candidates and how
they inuence their adoption of social media. e rst characteristics
I explore relate to the demographics of the candidates, specically age,
gender, and education level. e second characteristic is awareness of
social media, while the third is years of experience with social media.
ese characteristics inuence how early a candidate is likely to
adopt social media, and marks them out as either early adopters, early
majority, late majority, or laggards. Innovators are the rst wave to
try a new technology, followed by early adopters and then the early
majority. e late majority are slower to adapt to new technology while
laggards, who are the last adopters, are most resistant to change and see
a high risk in adopting new technology.
CANDIDATE DEMOGRAPHICS
I examine candidates’ age, gender, and education level and how these
factors impact their rate of adoption of social media.
10. Ibid., 16.
11. Foot and Schneider, Web Campaigning, 170–175.
Age
Most of the candidates who campaigned in the 2013 elections
belonged to a younger age demographic. Results indicated that the
greater proportion of candidates were below 50 years of age (see Table
3.1). Younger candidates were quicker to adopt social media than older
candidates (see Table 3.2). For example, younger candidates (21–35
years) at the gubernatorial level posted 100 percent adoption. A similar
category posted 83.3 percent adoption at the senatorial position, 71.4
percent at the women representative position, and 93.3 percent at the
parliamentary level. However, chi-square results (P>0.05) indicate that
there was no signicant variation across the ages.
e implication is that the digital divide occasioned by age continues
to diminish in the adoption of social media for election campaigns. e
results agree with ndings from a study conducted by Williams and
Gulati which indicated that age demarcates an important, albeit di-
minishing, digital divide in social networks more generally, and much
more so in the use of Facebook in particular.12
What is interesting is to remember that the age factor might have
been mitigated by older candidates who employed managers for their
social media sites. In an interview conducted with a political candidate
running for the position of women representative, whose own age was
above 51 years, she stated that she had a university student running her
online campaign:
“ere was a young man who actually came to me and said,
‘Mama, I think if you use Facebook you will be able to reach
more people. So, I was encouraged to use Facebook by a stu-
dent from Moi University.”
For this candidate, it did not matter that she had limited knowledge
of social media use for political communication, because she procured
an assistant who could help run her online campaign. is observa-
tion is supported by ndings from a study conducted by Williams and
12. Williams and Gulati, “Social Networks in Political Campaigns,” 66.; Gulati and
Williams, “Social Media and Campaigns 2012,” 584.
48
SOCIAL MEDIA & POLITICAL CAMPAIGNS IN KENYA | John O. Ndavula
49
Chapter 3 Diffusion of Social Media in the 2013 Election Campaigns
Gulati that indicated that the importance of age as a determinant of
social network usage is diminishing.13
However, we observed other candidates who were averse to using
social media strategies for their campaigns and saw a high risk in utiliz-
ing them. One parliamentary candidate aged above 51 years observed
that:
“On social media, there is anonymity which creates an ambigu-
ous context. You are not sure who the person you are engaging
with is. e older generation prefers to think it [social media]
is neither here nor there.”
Gender
e Constitution of Kenya 2010 proposes gender parity in elections.
In chapter 6 of the Constitution, which deals with the representation
of people, one of the general principles expressed is to ensure fair rep-
resentation of women.
Results show that whereas both genders actively participated in the
2013 elections, there were more male candidates (75 percent) par-
ticipating in the electioneering than females (25 percent). When the
gender demographic was pitted against social media adoption, results
indicated that more male candidates (88.2 percent) used social media
for political communication than female candidates did (86.1 percent)
(see Figure 3.2). ere were high levels of social media utilisation by
both genders, however, implying that gender dierences in adoption of
social media for election campaigns were diminishing.
I then examined the candidates’ gender and social media adoption
rates as they related to the level of oce being campaigned for by the
candidate (see Table 3.3). Results showed that 100 percent of male
candidates at the presidential level used social media. At the guber-
natorial level, 100 percent of female candidates used social media, as
compared to 92 percent of male candidates. At the senatorial level,
85 percent of male candidates used social media as compared to 80
13. Ibid., 584.
percent of women. At the women representative level, 85 percent of the
women used social media. At the parliamentary level, more women
than men used social media as indicated by 89 percent and 84 percent
respectively. None of the chi-square results for social media use for
various titles of oce vied for were signicant (P>0.05), implying that
the variation across gender was not statistically signicant.
Although most candidates from both genders used social media,
those at the higher levels of oce used it more than those at the lower
levels of oce. It follows that candidates for the presidential, guberna-
torial, and senatorial levels of oce represented larger constituencies;
hence they needed to reach a greater majority of voters.
Although some studies contend that women feel less comfortable
with new information technology, these ndings suggest no signicant
dierence in the adoption of social media for political communication
Figure 3.2: Politicians’ Adoption of Social Media
50
SOCIAL MEDIA & POLITICAL CAMPAIGNS IN KENYA | John O. Ndavula
51
Chapter 3 Diffusion of Social Media in the 2013 Election Campaigns
across gender.14 is likely means that as internet and computer usage
becomes widespread, the gap accounting for usage between genders
narrows.15 In sum, there are indications of a narrowing gender gap in
social media adoption among political candidates in Kenya.
Education Level
Most candidates who vied for various posts in the 2013 general
elections had a tertiary level of education. A greater proportion of
candidates were degree holders (see Table 3.4). e trend towards hav-
ing a political class that is better educated can possibly be attributed
to a campaign, embedded in the Constitution of Kenya 2010, to have
a minimum tertiary level educational threshold for contestants vying
for various posts. Although not successful, this push can be seen to
have encouraged many candidates to have better academic credentials.
When education was pitted against social media adoption, results in-
dicated that candidates with university degrees tended to adopt social
media more than those without (see Table 3.5). Indeed, higher levels of
education tend to make people more comfortable with, and skilled in,
the use of technology.16
AWARENESS
Information was sought on the candidates’ awareness of social me-
dia use for election campaigns. In general, the results show that most
political candidates were aware of the use of social media for election
14. Joel Cooper and Kimberlee D. Weaver, Gender and Computers: Understand-
ing the Digital Divide (Mahwah, NJ: Princeton University, 2003), 13–14.; Auren
Homan, “e Social Media Gender Gap.” Accessed January 23, 2015. http://www.
bloomberg.com/bw/stories/2008-05-19/the-social-media-gender-gapbusinessweek-
business-news-stock-market-and-nancial-advice; Ira M. Wasserman and Marie
Richmond-Abbott, “Gender and the Internet: Causes of Variation in Access, Level,
and Scope of Use,” Social Science Quarterly, 86:1 (2005), 252–270.
15. Jan van Dijk, e Deepening Divide (ousand Oaks, CA: Sage, 2005), 61.
16. Williams and Gulati, “Social Networks in Political Campaigns,” 66.; Gulati and
Williams. “Social Media and Campaigns 2012,” 579.
campaigns (see Figure 3.3). Awareness at the presidential level was 100
percent, at the gubernatorial level 96 percent, at the senatorial level
96.8 percent, at the women representative level 90.9 percent, and at the
parliamentary level 95.2 percent. e high awareness corresponds to
the high adoption levels of social media (see chapter 2).
Results indicated that although both genders posted a high level of
awareness, men showed a marginally higher level of awareness (see
Figure 3.4). Regardless of gender, candidates at higher levels of oce
were relatively more aware of social media use for election campaigns
than those at lower levels of oce (see Table 3.6). Women candidates
running for higher levels of oce were more aware than those at the
lower levels of oce: 100 percent of gubernatorial candidates, 100
percent of senatorial candidates, 90 percent of women representative
candidates, and 88.6 percent of parliamentary candidates. Male candi-
dates exhibited relatively similar levels of awareness across the levels
Figure 3.3: Candidates’ Awareness of Social Media Use
52
SOCIAL MEDIA & POLITICAL CAMPAIGNS IN KENYA | John O. Ndavula
53
Chapter 3 Diffusion of Social Media in the 2013 Election Campaigns
of oce. e variation of awareness across gender was not statistically
signicant (P>0.05). Although, traditionally, men are the early adop-
ters of new technologies, these ndings indicated a normalization of
adoption rates in the context of election campaigns. 17
YEARS OF INTERNET EXPERIENCE
Information was sought on the candidates’ years of internet experi-
ence and how it related to social media adoption. Results obtained from
the chi-square test were (X2 = 24.293, df = 12, Sig = 0.019) indicating
that the variation in the years of internet experience was signicant
(see Table 3.7). Candidates campaigning for higher electoral oces
had more internet experience than those campaigning for lower of-
ces. Women representatives, for that matter, had the least internet
experience.
17. Homan, “e Social Media Gender Gap.”
I investigated how the internet experience gap impinged on the
adoption of social media for election campaigns. Data in Table 3.8
indicated that, among candidates with more than ten years’ internet
experience, 89.8 percent of them used social media. Among candidates
with three to ten years’ experience with the internet, 92.5 percent of
them used social media. Among candidates who had less than three
years’ experience, 83.7 percent of them used social media. Finally, 33.3
percent of candidates with no internet experience used social media.
Results obtained from the chi-square test were (X2 = 49.194, df = 3, Sig
= 0.000) indicating that the variation between years of internet experi-
ence and social media use was signicant. It follows that candidates
who had more than three years of internet experience were more likely
to use social media than those who had fewer years of internet experi-
ence.
Technological Factors and Social Media Adoption
Technological characteristics are key to understanding how candi-
dates adopt social media for election campaigns. e characteristics
I examine are relative advantage, compatibility, complexity, and
observability. Relative advantage is the degree to which social media
are perceived as better in terms like ease of use, aordability, control,
and convenience. Complexity is the degree to which social media are
relatively dicult to understand and use. Observability is the extent
to which candidates see the results of social media for election cam-
paigns. Compatibility relates to how social media are consistent with
past experiences and needs of the candidates.
RELATIVE ADVANTAGE
Relative advantage is the degree to which social media are perceived
as better than other media. Most candidates felt that social media af-
forded them many advantages in reaching voters (see Figure 3.5). A
slight majority of candidates (60 percent) felt that they could control
their online communication with voters. Others cited the low costs
Figure 3.4: Distribution of Gender and Awareness of Social Media
54
SOCIAL MEDIA & POLITICAL CAMPAIGNS IN KENYA | John O. Ndavula
55
Chapter 3 Diffusion of Social Media in the 2013 Election Campaigns
of campaigning on social media. Indeed, with online communication,
the cost of information retrieval and communication in general falls
and political participation becomes less costly. Since these costs are
extremely low, most candidates see the internet as a fundamental
component of any communication and mobilization strategy.18 Other
candidates cited access to computers and adequate internet speeds as
a prerequisite to successful online campaigning. In other words, they
observed that slow internet speeds hampered communication and
stood in the way of adoption of social media. From the interviews
conducted, candidates stated that posting content, especially photos
and videos, posed a challenge when internet speeds were slow.
e positive aspects noted above account for the higher adoption
rates of social media by Kenyan political candidates. Diusion of inno-
vations literature states that a new technology has a higher chance for
rapid adoption if it is compatible with the needs of potential adopters.
18. Philip N. Howard, New Media Campaigns and the Managed Citizen (New York:
Cambridge University Press, 2005), 1–69.; Williams and Gulati, “Social Networks in
Political Campaigns,” 55.; Gulati and Williams, “Social Media and Campaigns 2012,”
579.
COMPLEXITY
Complexity is the degree to which social media are relatively dicult
to understand and use. Most of the candidates (77 percent) perceived
social media as easy to use, a factor which accounts for the high inci-
dence of usage. Diusion of innovations theory states that complexity
of innovations is negatively correlated with the rate of adoption.19
When asked whether they required special training to leverage so-
cial media better, over half of the candidates objected. One member of
a presidential social media campaign team stated that:
“Social media doesn’t require that much training. It [sic] is just
like a normal email account, but here you have more leeway
to upload photos, update statuses…so I think it is not that dif-
cult to use social media. It doesn’t really need training per se
so long as you can follow instructions.”
is statement aptly captures the view candidates have that social
media are accessible and easy. However, utilizing social media ef-
fectively for election campaigns candidates may require training (see
chapter 4). Scholars observe that social networks benet from many
members or from serving a specic niche, and they require technical
expertise and specialised knowledge to be designed and leveraged ef-
fectively.20
In support of this view, a social media site manager for a gubernato-
rial candidate stated:
“I would say training is a must because there are advanced
ways of using Facebook. You can use hashtags, at-symbols to
make your posts more visible. You must know how to do catchy
headings without spaces in between words but capitalising
each new word for visibility purposes. You also must gather
what’s trending so that you can piggy back on it. It is only with
training that politicians can become strategic users of social
media. Again, if you want to target groups and ensure a wider
reach, you must be strategic in your use. You can, for instance,
19. Rogers, Diusion of Innovations, 16.
20. Williams and Gulati, “Social Networks in Political Campaigns,” 54.
Figure 3.5: Factors Inuencing Adoption of Social Media
56
SOCIAL MEDIA & POLITICAL CAMPAIGNS IN KENYA | John O. Ndavula
57
Chapter 3 Diffusion of Social Media in the 2013 Election Campaigns
tag a lot of people, you can target already formed groups. For
instance, if you are campaigning in Kiambu, you have avail-
able neighbourhood groups like ‘Kiambu Yetu,’ ‘United States
of Kiambu,’ or ‘Kiambu Ndio Home.’ Such groups pool together
a community of users, some of whom grew up together and
therefore have a strong bond. Since you cannot come to them
physically, social media become a place where you can connect
with them.”
Apart from Facebook, Twitter also requires expertise. O’Connor et
al. state that social media in general and Twitter in particular require
technical expertise and specialised knowledge in order to be leveraged
eectively, noting that the micro-blogging nature of Twitter makes it
challenging to use. 21
Subsequently, most candidates felt that having social media site
managers was important. A presidential candidate explained:
“Having a social media team is important. When campaign-
ing on Twitter for example and you attend a rally, there are
certain points you want people out there to know. ese can
include policies, declarations and so on. It is important to have
someone who posts the messages on social media as I deliver
my speech. I cannot give a speech and tweet at the same time.”
Further, a social media site manager for a presidential candidate stated:
“…once you decide to use it [social media] as your campaign
tool the dynamics change. It is no longer the social media plat-
form you use for just communicating and sharing on a small
scale. It means you will have responses and questions in big
numbers because you are targeting a big market. erefore,
the candidate cannot handle this alone. ey needed input
of professional sta that can handle the communication. is
would mean sta who are both technical, and who are trained
in communications or public relations to handle matters.”
21. O’Connor et al., “From Tweets to Polls: Linking Text Sentiment to Public
Opinion Time Series.” Accessed December 9, 2016. http://www.aaai.org/ocs/index.
php/ICWSM/ICWSM10/paper/viewFile/1536/1842
However, one gubernatorial candidate with dissenting views stated:
“Sta can manage your website but not Facebook and Twitter.
ere are some personal things that you would need to explain.
Maybe have a personal conversation with your voters. Some
comments may be so frank as to require your personal input.
Your team may be doing public relations work, by siing and
editing the comments so that you don’t see the bigger picture.
You need to know if there is growing propaganda against
you so that you can respond to it. You need to know what is
happening on the ground. You need to know who is spreading
negative propaganda, you need to put faces to comments so
that you respond to them. It is important for politicians to get
personal on social media.”
OBSERVABILITY
Observability is the extent to which candidates see the results of
social media for election campaigns. Most candidates felt that social
media allowed them to monitor and evaluate the progress of their elec-
tion campaigns (see Table 3.9). However, candidates who showed least
agreement were women representatives.
Interviews with political candidates and sta on their social media
campaign teams supported this view. For instance, one presidential
social media site manager stated:
“It [social media] is the best thing that ever happened to
people campaigning. In the days past, candidates would use
posters, loudspeakers on cars, radio and TV but they never
got feedback. For this one [social media] feedback is instant. If
you were campaigning in Nairobi County, you would know the
response coming from Dagoretti Sub County, Makadara Sub
County and so on. You would know whom to target and whom
not to target, therefore social media was [sic] very eective.”
A senatorial candidate stated that:
“I corrected a few things that had gone wrong on my campaign
because of social media. For example, if someone made a
comment that said I had never visited their area, yet I had
promised to visit, I would update my schedule to include a
58
SOCIAL MEDIA & POLITICAL CAMPAIGNS IN KENYA | John O. Ndavula
59
Chapter 3 Diffusion of Social Media in the 2013 Election Campaigns
visit to that area. I could go there aer one week because I
don’t want to lose that one person. I could also head to an area
to correct something that someone bad has said, especially if
I had sent an aide there who misrepresented me, just because
someone has given me direct feedback. Social media is [sic]
good because some people may not be courageous enough to
face me directly and tell me this wasn’t done right. However, it
is easy for them to go on social media because there they can
use pseudo names and bravely tell the truth. e can say, “is
is what is happening, this is what could have been done, this is
what you need to do if you really need to win.”
In another interview, a parliamentary candidate observed that:
“…anybody who is making a comment represents a particular
constituency not based on boundaries, but by the particular
view they represent. Sometimes they are accurate, other times
they are not. Social media can give politicians some direction
on whether they are doing well or whether they are not doing
well. However, it depends on the context. In urban areas views
on social media would have more impact because many peo-
ple would have a link to that post. In rural areas, they would
b e f e w .”
On the contrary, a parliamentary candidate observed that:
“Most politicians get information through other means other
than social media. I can tell you they have their people on the
ground. Politicians do not heavily rely on social media to know
their rating.”
Taken together, the evidence from the foregoing analyses provides
support that social media play an important role in campaign monitor-
ing and evaluation. Social media provide candidates with tools to trace
public opinion during an election campaign and to assess the impact
of their campaign. Among studies that support this view are those
conducted by Kavanagh and Scammell, which identify monitoring and
evaluation as a key component of political campaigns. 22
COMPATIBILITY
Compatibility relates to how social media are consistent with past ex-
periences and needs of the candidates. Most candidates felt that social
media were compatible with their past online experience, noting that
using them did not necessarily require training. Other candidates cited
ease of communication online as working to their advantage. Further
still, some candidates reported experiencing a sense of interpersonal
connection with voters online. Some reported that conversation on
Facebook led to the establishment of a relationship with voters.23 is
is comparable with candidates’ experiences oine as well. Essentially,
candidates want to use campaign methods that endear them to voters
such as rallies or door-to-door campaigns. Network society theory
states that political actors interacting on social media are engaging
in a relationship. e theory further posits that the internet creates
connections, and that the interactants have the potential to create new
values from these connections. Further, all interactions online create
a collective mental experience that is virtual, and the virtuality is a
fundamental dimension of everybody’s reality.24 In other words, the
relationships between candidates and voters depend on what happens
in this media-centred communication space. Findings from studies
conducted by Williams and Gulati indicate that candidates using social
media establish a relationship with voters.25
22. Dennis Kavanagh, “New Campaign Communications: Consequences for Po-
litical Parties,” Harvard International Journal of Press and Politics 1:3 (1996): 60–76.;
Dennis Kavanagh, Election Campaigning: e New Marketing of Politics (Oxford:
Blackwell Publishers, 1995).
23. Williams and Gulati, “Social Networks in Political Campaigns,” 53.
24. Castells, “Informationalism, Networks, and the Network Society,” 46.
25. Williams and Gulati, “Social Networks in Political Campaigns,” 53.
60
SOCIAL MEDIA & POLITICAL CAMPAIGNS IN KENYA | John O. Ndavula
61
Chapter 3 Diffusion of Social Media in the 2013 Election Campaigns
Even then, a signicant proportion of candidates (35 percent) do not
feel interpersonally connected to voters, and this had a negative eect
on adoption rates. In an interview, a women representative candidate
stated that:
“…I don’t know those faces; some of them don’t know me.
ey have never met me. So, I may not believe what they are
talking about.”
us, we see that, while most candidates found social media easy to
use, highly compatible with their experience, and useful as a tool for
connecting with voters on a personal level, a small minority of candi-
dates still expressed some reservations about social media adoption.
Environmental Factors and Social Media Adoption
Environmental factors have an impact on a candidate’s adoption de-
cisions. e environment is the social system into which an innovation
is being introduced. I examined the level of oce, competitiveness of
the race, constituency type, voter demographics, and party aliation
as aspects of the environment in which social media were introduced.
LEVEL OF OFFICE
e 2010 Constitution of Kenya separated power and dispersed it
vertically in terms of the dierent levels of oce, namely presidential,
gubernatorial, senatorial, women representative, and parliamentary.
Findings showed that level of oce had little inuence on adoption
decisions (see Figure 3.6). In other words, most candidates did not
feel obliged to use social media because of the level of oce they were
campaigning for. is is a surprising nding given that some scholars
point out that level of oce is a strong determinant of social media
use.26 Indeed, while some candidates felt that level of oce was least
inuential, other candidates felt that its impact was signicant. One
senatorial candidate interviewed observed that:
26. Foot and Schneider, Web Campaigning, 171–172.
“A member of county assembly might choose not to use
social media because the campaign ground to be covered is
comparatively small. He can choose to perform a door-to-door
campaign. e same applies to a member of parliament who
can hop from one meeting to another, from one church to
another and so on. But if you want to campaign in nine or ten
constituencies for a governor or senator, you might not reach
everyone before the close of campaign time. So, social media
becomes useful. erefore, the level of oce you are vying for
is a consideration for engaging social media.”
Figure 3.6: Level of Oce as a Determinant of Social Media Adoption
62
SOCIAL MEDIA & POLITICAL CAMPAIGNS IN KENYA | John O. Ndavula
63
Chapter 3 Diffusion of Social Media in the 2013 Election Campaigns
COMPETITIVENESS OF THE RACE
Nearly half of the candidates indicated that the competitiveness of
the race inuenced their decision to adopt social media for election
campaigns (see Figure 3.7). Similarly, ndings indicate that in US elec-
tions the competitiveness of the race contributed to adoption of online
campaigns.27
CONSTITUENCY TYPE
Constituencies were categorised as either urban, peri-urban, or ru-
ral. Results indicate that most of the constituencies (48.7 percent) were
rural, followed by peri-urban (31 percent) and urban (20.3 percent)
(see Figure 3.8).
27. Paul. S. Herrnson, Atiya K. Stokes-Brown, and Mathew Hindman. “Campaign
Politics and the Digital Divide: Constituency Characteristics, Strategic Considera-
tions, and Candidate Internet Use in State Legislative Elections,” Political Research
Quarterly 60:1 (2007): 31–42.; Elaine C. Kamarck, “Political Campaigning on the
Internet: Business as Usual?” in Governance.Com: Democracy in the Information Age.
ed. Elaine C. Kamarck, and Joseph S. Nye (Washington, DC: Brookings Institution,
2002), 81–103.
However, there were no major dierences in the adoption of social
media across the constituency types (see Figure 3.9). Although can-
didates targeting peri-urban voters showed a marginally higher usage
of social media (92.2 percent), results from chi-square tests showed
that the variation across the constituency types was not statistically
signicant (X2 = 1.375, df = 2, Sig = 0.503). Although diusion of in-
novations theory postulates that adoption decisions are dependent
on characteristics of the environment, which according to Foot and
Schneider includes particular constituency types, it is evident that this
had little inuence on social media adoption decisions in Kenya.28 is
could be explained by the fact that voters in rural areas are experienc-
ing increasing access to the internet through mobile phones. Research
indicates a mobile phone penetration rate of 78 percent, and that
28. Foot and Schneider, Web Campaigning, 171–172.; Ward and Gibson, “Euro-
pean Political Organisations and the Internet,” 38.
Figure 3.7: Competitiveness of the Race as a Determinant of Social Media Adoption
Figure 3.8: Electoral Constituency Types
64
SOCIAL MEDIA & POLITICAL CAMPAIGNS IN KENYA | John O. Ndavula
65
Chapter 3 Diffusion of Social Media in the 2013 Election Campaigns
34 percent of the Kenyan population access the internet via mobile
phones.29 Although urban areas have greater internet access than rural
areas, the dierences are diminishing.
VOTER DEMOGRAPHICS
Voter demographics examined include age of constituents, income
level, and literacy levels and how they inuenced candidates’ decisions
to adopt social media for election campaigns.
29. Humanipo, “Kenya ICT Board Launches Julisha ICT Survey Report 2013.”
Accessed September 3, 2016. http://icta.go.ke/pdf/Julisha%20Final%20Report%201.
pdf; Communications Commission of Kenya, “Mobile Penetration in the Country
Continues to Increase.” Accessed November 6, 2016. http://www.cck.go.ke/mobile/
news/index.html?nws=/news/2013/Mobile_penetration.html
Age of Constituents
Most candidates used social media to target young adult voters (see
Figure 3.10). Results indicated that 66.1 percent of candidates targeted
voters aged between 18 and 25 years, while 18.9 percent of candidates
targeted voters aged between 26 and 35 years. A further 3.6 percent of
candidates targeted voters aged between 36 and 50 years, while only
1.6 percent targeted voters above 51 years.
Scholars posit that social networking sites are well suited for send-
ing tailored campaign messages to specic voter groups.30 Literature
from social marketing theory points to the need to target messages at
audience segments most receptive or susceptible to those messages.
By identifying the most vulnerable segments and then reaching them
with the most ecient channel available, targeting strategies reduce
promotional costs while increasing eciency.31 Political candidates in
Kenya are seen, therefore, as harnessing the power of social media to
reach out to young voters using the most ecient channel available to
them.
Most candidates agree that the average age of constituents informed
their social media adoption decisions. Candidates felt inclined to use
social media if they considered most of their likely voters to be drawn
from a younger demographic (see Table 3.10). is view was supported
by political candidates and sta on their social media campaign teams.
For instance, one presidential campaign sta member observed:
“When deciding to use a medium for our campaigns, we used
media accessible to our target group. For information on youth
empowerment, or youth mobilizing and employment, we used
30. Jody Baumgartner and Jonathan Morris, “Myfacetube Politics: Social Network-
ing Websites and Political Engagement of Young Adults,” Social Science Computer
Review, 28:24 (2010): 24–44.; Mathew Kushin and Masahiro Yamamoto, “Did Social
Media Really Matter? College Students’ Use of Online Media and Political Decision
Making in the 2008 Election,” Mass Communication and Society, 13 (2010): 608–630.;
Sonja Utz, “e (Potential) Benets of Campaigning Via Social Network Sites,” Jour-
nal of Computer Mediated Communication, 14 (2009): 221–243.
31. Baran and Davis, Mass Communication eory, 280.
Figure 3.9: Constituency Type as a Determinant of Social Media Adoption
66
SOCIAL MEDIA & POLITICAL CAMPAIGNS IN KENYA | John O. Ndavula
67
Chapter 3 Diffusion of Social Media in the 2013 Election Campaigns
social media. But if the piece of information didn’t concern the
youth, we avoided social media because the impact wouldn’t
be much. ere were voters above 50 years who know about
Facebook, for instance, but they wouldn’t go there to get cam-
paign information. e youth don’t even need to go to those
sites since they are always there, alerts keep popping up on
their devices.”
ese observations agree with ndings from a study conducted by
Xenos and Bennett, which found age to be an important determinant
of social media usage and adoption.32
Income Level of Constituents
Most candidates observed that income level of the constituents was a
determinant for social media adoption (see Table 3.11). e chi-square
results were (X2 = 20.26, df = 16, Sig = 0.209) suggesting that there was
no signicant variation on this opinion across the various categories of
political candidates. Diusion of innovations literature suggests that
32. Michael A. Xenos and Lance W. Bennett, “e Disconnection in Online Poli-
tics: e Youth Political Web Sphere and US Election Sites, 2002-2004,” Information,
Communication, & Society 10:4 (2007): 443–464.
constituency factors, such as income levels, would indeed inuence
adoption rates, so broad agreement among candidates on this aspect is
unsurprising.33
Literacy Levels of Constituents
Most candidates agreed that literacy level of constituents was a de-
terminant for social media adoption (see Table 3.12). e chi-square
results were (X2 = 6.661, df = 12, Sig = 0.879), suggesting that there was
no signicant variation on this opinion across the various categories
of political candidates. ese ndings agree with conclusions made by
Foot and Schneider, who posited that determiners of adoption in the
context of political communication draw on factors such as literacy
level of constituents.34
PARTY AFFILIATION
Political parties are the backbone of any democracy (see chapter 1).
In the 2013 elections, dierent political parties sponsored various can-
didates. Overall, most candidates participating in the elections came
from the major political parties. is implies that the major parties are
still dominating elections in Kenya, and most contestants align them-
selves with these parties to clinch political seats. Findings showed that
41.9 percent of candidates were inuenced by their political parties
to use social media (see Figure 3.11). is shows that political parties
have some inuence on use of media for election campaigns. Political
parties adopted social media regardless of their party strength or posi-
tion. For example, although a party like TNA ran on a digital campaign
platform, candidates did not feel pressured by their parties to adopt
online campaigns. is implies that all parties exhibited similar behav-
iour in terms of adoption of social media.
33. Ward and Gibson, “European Political Organisations and the Internet,” 25–39.
34. Foot and Schneider, Web Campaigning, 171–172.
Figure 3.10: Target Audience for Social Media Messages
68
SOCIAL MEDIA & POLITICAL CAMPAIGNS IN KENYA | John O. Ndavula
69
Chapter 3 Diffusion of Social Media in the 2013 Election Campaigns
Determinants of Social Media Use for
Election Campaigns in Kenya
Diusion of innovations theory suggests that the primary deter-
minants of adoption of a new technology encompass adopter charac-
teristics, technological characteristics, and environmental factors in
general. It was important, therefore, to attempt to single out factors
specic to social media adoption by candidates in election campaigns
in Kenya. I used a binary logistic regression model to determine the
factors critical to adoption (see Table 3.13). I conducted a correlation
test to ensure that no variable eect was duplicated (see Table 3.14).
Both positive and negative correlations existed between the vari-
ables, with the magnitude of the correlations ranging from very weak
to average. e highest magnitude was 0.578, indicating that there
were no very strong correlations between the variables included in
the modelling. At 95 percent condence level, the p-value was set at
0.05. Findings indicate that the variables with p-values < 0.05 were:
familiarity with technology (p-value = 0.000), political party inuence
(p-value = 0.036), party aliation (p-value = 0.024), years of internet
use (p-value = 0.016), and social media awareness (p-value = 0.000)
(see Table 3.15). e other variables, referenced previously, did not
make statistically signicant contributions to the model.
e logistic regression model was statistically signicant: χ2 = 80.639,
p-value < 0.000. e model explained 75.4 percent (Nagelkerke R2) of
the variance in social media use in election campaigns in Kenya and
correctly classied 92.1 percent of cases. e ndings indicate that
familiarity with technology was 4.602 times more likely to inuence
the adoption of social media. Increasing familiarity with technology,
awareness of the use of social media for election campaigns, and years
of internet use were all associated with an increased likelihood of
adopting social media. However, increasing political party inuence
and party aliation was associated with a reduction in the likelihood
of adopting of social media.
is is a strong model, because 75.4 percent of the variations were de-
termined by the variables within the model. e model could therefore
be used to predict candidates’ behaviour during political campaigns,
as far as social media adoption is concerned. When dealing with the
model, it should be noted that 24.6 percent of the variations in social
media use for election campaigns in Kenya are explained by variables
outside the model. Further investigation is necessary to identify these
remaining factors.
Conclusion
is chapter has identied some of the factors upon which adop-
tion decisions depend: candidate characteristics, technological factors,
and characteristics of the environment. Regarding characteristics
of candidates, it was found that demographics, awareness of social
media as a strategy, and years of internet experience were all relevant
to adoption decisions. Personal attributes such as higher education
Figure 3.11: Political Party Inuence and Social Media Adoption
70
SOCIAL MEDIA & POLITICAL CAMPAIGNS IN KENYA | John O. Ndavula
71
Chapter 3 Diffusion of Social Media in the 2013 Election Campaigns
levels and relatively younger age could account for the rise in adoption
among candidates. A high level of awareness of social media coupled
with years of internet usage among the candidates was shown to have a
signicant inuence in adoption of social media.
Technological attributes such as relative advantage, complexity,
observability, and compatibility also account for the steeper adoption
rates. Most candidates found social media easy to use, while others said
that they were consistent with their political communication needs.
Candidates stated that social media eased their communication with
voters. is therefore could account for the higher adoption rates of
social media by Kenyan political candidates. Indeed, diusion of in-
novations theory posits that a new technology has a higher chance for
rapid adoption if it is compatible with the needs of potential adopters,
and social media have proven to be a powerful and convenient com-
munication tool for a new crop of Kenyan politicians.
e cost of updating proles and posting content on social media
sites was extremely low, a factor that encouraged candidates to employ
the online campaign as a fundamental component of their communi-
cation and mobilisation strategy. is implies that the cost of democ-
racy may come down if social media strategies are implemented on a
large scale and utilised optimally. However, interviews with candidates
established that most them were in favour of hiring social media site
managers. is raises the possibility that professionalising social media
usage may transfer some of the inequalities currently seen with mass
media campaigns onto social media. It is likely that better-nanced
candidates would be able to utilize social media more eectively than
less-endowed candidates would.
Results from this study give support to the notion that constituency
characteristics indeed inuence social media adoption for election
campaigns. Characteristics such as level of oce, competitiveness of
the race, constituency type, voter demographics, and party aliation
were likely to inuence social media adoption.
e regression model suggested that increasing familiarity with tech-
nology, awareness of the use of social media for election campaigns,
and years of internet use were associated with an increased likelihood
of adopting social media use. However, increasing political party in-
uence and party aliation were associated with a reduction in the
likelihood of adopting of social media use. e explanatory power of
the model requires more systematic and extensive replication before it
can be employed to adequately explain adoption behaviour.
Overall, this chapter has established factors that contribute to the
adoption of social media in Kenya. ese social media sites provide ac-
cessible means for candidates to campaign online, and it is incumbent
upon the candidates and online campaign strategists to recognize this
potential and utilise these platforms eectively.
72
SOCIAL MEDIA & POLITICAL CAMPAIGNS IN KENYA | John O. Ndavula
73
Chapter 3 Diffusion of Social Media in the 2013 Election Campaigns
Table 3.1: Candidates’ Age
Electoral Ofce (%)
Age (n = 307) President Governor Senator Women Rep. MP Chi-square test
> 20 years 0 0 3 3 1
X2 = 7.575
df = 12
Sig = .817*
21 – 35 years 0 20 23 21 29
36 – 50 years 33 36 52 55 50
51+ years 67 44 26 18 20
* Correlation is signicant at the 0.05 level.
Table 3.2: Distribution of Candidates by Age and Social Media Adoption
Electoral Ofce
(%)
Age Chi-square
test
Below 20
years
21 – 35
years
36 – 50
years
51+
years
President
(n = 6)
Yes 0.0 0.0 100.0 100.0 a
No 0.0 0.0 0.0 0.0
Governor
(n = 23)
Yes 0.0 100.0 88.9 88.9 X2 = 0.661
df = 2
Sig = 0.719*
No 0.0 0.0 11.1 11.1
Senator
(n = 29)
Yes 100.0 83.3 86.7 85.7 X2= 0.194
df = 3
Sig = 0.978*
No 0.0 16.7 13.3 14.3
Women-
Rep.
(n = 33)
Yes 0.0 71.4 94.4 71.4 X2= 6.866
df =
Sig = 0.076*
No 100.0 28.6 5.6 28.6
MP
(n = 205)
Yes 0.0 93.3 86.3 86.0 X2= 1.010
df = 2
Sig = 0.604*
No 0.0 6.7 13.7 14.0
a No statistics are computed because the variable is a constant.
* Correlation is signicant at the 0.05 level.
Table 3.3: Relationship between Gender and Social Media Adoption
Title of
ofce
Use of
social
media
Gender
Chi-square
test
Male Female
Freq. % within
gender Freq. % within
gender
President Yes 6 100 0 0 a
Governor
Yes 22 92 1 100 X2 = 0.179
df = 1
Sig = 0.857*
No 2 8 0 0
Senator
Yes 22 85 4 80 X2 = 0.048
df = 1
Sig = 0.627*
No 4 15 1 20
Women
Rep.
Yes 0 0 28 85 X2 = 1.616
df = 1
Sig =0.320*
No 0 0 5 15
MP
Yes 148 84 32 89 X2 = 0.002
df = 1
Sig = 0.616*
No 28 16 4 11
a No statistics are computed because the variable is a constant.
* Correlation is signicant at the 0.05 level.
74
SOCIAL MEDIA & POLITICAL CAMPAIGNS IN KENYA | John O. Ndavula
75
Chapter 3 Diffusion of Social Media in the 2013 Election Campaigns
Table 3.4: Candidates’ Education Levels
Level of Education Chi-square
test
Diploma Bachelor’s Master’s Doctorate
Electoral
Ofce (%)
President
(n = 6) 0.0 100.00 0.0 0.0
X2 = 14.685
df = 12
Sig = 0.259*
Governor
(n = 25) 0.0 68.00 24.00 8.00
Senator
(n = 31) 12.90 54.80 22.60 9.70
Women _
Rep. (n = 33) 33.30 45.50 21.20 0.0
MP
(n = 201) 23.90 55.70 16.90 3.50
* Correlation is signicant at the 0.05 level
Table 3.5: Candidates’ Education Level and Social Media Adoption
Electoral
Ofce (%) Level of Education Chi-square test
Diploma Bachelor’s Master’s Doctorate
President
(n = 6)
Yes - 100.0 - - a
No - - - -
Governor
(n = 25)
Yes - 88.2 100.0 100.0 X2 = 0.839
df = 2
Sig = 0.657*
No - 11.8 0.0 0.0
Senator
(n = 31)
Yes 100.0 76.5 85.7 100.0 X20= 1.805
df = 3
Sig =0.614*
No .0 23.5 14.3 0.0
Women
Rep.
(n = 33)
Yes 72.7 93.3 71.4 - X2 = 2.012
df = 2
Sig = 0.366*
No 27.3 6.7 28.6 -
MP
(n = 197)
Yes 73.9 94.5 82.4 100.0% X2 = 7.438
df = 3
Sig = 0.059*
No 26.1 5.5 17.6 0.0%
a No statistics are computed because the variable is a constant.
* Correlation is signicant at the 0.05 level.
Table 3.6: Relationship between Gender and Awareness of Social Media
Electoral Ofce Gender (%)Chi-square test
Male Female
President
(n = 6)
Yes 100.00 0.00 a
No 0.00 0.00
Governor
(n = 23)
Yes 95.20 100.00 X2 = 0.083
df = 1
Sig = 0.929*
No 4.80 0.00
Senator
(n = 31)
Yes 96.20 100.00 X2 = 0.207
df = 1
Sig = 0.833*
No 3.80 0.00
Women Rep.
(n = 33)
Yes 0.00 90.0 X2 = 0.166
df = 1
Sig = 0.860*
No 0.00 10.00
MP
(n = 209)
Yes 96.60 88.60 X2 = 1.972
df = 1
Sig = 0.199*
No 3.40 11.40
a No statistics are computed because the variable is a constant.
* Correlation is signicant at the 0.05 level.
Table 3.7: Candidates’ Years of Internet Experience
Electoral Ofce (%) Chi-
square
test
President
(n = 6)
Governor
(n = 25)
Senator
(n = 31)
Women Rep.
(n = 34)
MP
(n = 211)
Years of
Internet
Use
No
experience 0.0 8.0 9.7 11.8 2.8
X2 =
24.293
df = 12
Sig =
0.019*
Less than
3 years 0.0 0.0 25.8 35.3 12.8
3 – 10
Years 50.0 68.0 45.2 47.1 64.5
More than
10 years 50.0 24.0 19.4 5.9 19.9
* Correlation is signicant at the 0.05 level.
76
SOCIAL MEDIA & POLITICAL CAMPAIGNS IN KENYA | John O. Ndavula
77
Chapter 3 Diffusion of Social Media in the 2013 Election Campaigns
Table 3.8: Years of Internet Experience and Social Media Use
Years of Internet
Experience
Social Media Use (%) Chi-square test
Yes No
No experience 33.3 66.7
X2 = 49.194
df = 3
Sig = 0.000*
Less than 3 years 83.7 16.3
3–10 years 92.5 7.5
More than 10 years 89.8 10.2
* Correlation is signicant at the 0.05 level.
Table 3.9: Monitoring and Evaluation Using Social Media
Electoral Ofce (%) Chi-square
test
President
(n = 4)
Governor
(n = 25)
Senator
(n = 30)
Women Rep.
(n = 34)
MP
(n = 209)
Strongly
disagree 0.0 4.0 6.7 8.8 3.8
X2 = 8.780
df = 16
Sig = 0.922*
Disagree 25.0 16.0 13.3 23.5 13.9
Neither agree
nor disagree 0.0 16.0 16.7 11.8 17.7
Agree 25.0 48.0 36.7 44.1 43.1
Strongly
agree 50.0 16.0 26.7 11.8 21.5
* Correlation is signicant at the 0.05 level.
Table 3.10: Age of Constituents and Social Media Adoption
Electoral Ofce (%) Chi-square
test
President
(n = 6)
Governor
(n = 25)
Senator
(n = 28)
Women Rep.
(n = 34)
MP
(n = 207)
Disagree 0.0 0.0 7.1 5.9 2.9
X2 = 8.376
df = 16
Sig = 0.025*
Neither agree
nor disagree 0.0 4.0 7.1 14.7 4.8
Agree 100.0 76.0 78.6 64.7 75.4
Strongly
agree 0.0 20.0 7.1 14.7 16.9
* Correlation is signicant at the 0.05 level.
Table 3.11: Income Level of Constituents and Social Media Adoption
Electoral Ofce (%) Chi-square
test
President
(n=6)
Governor
(n=25)
Senator
(n=28)
Women Rep.
(n=33)
MP
(n=207)
Strongly
disagree 0.0 8.0 0.0 0.0 2.9
X2=20.26
df = 16
Sig = 0.209*
Disagree 0.0 4.0 14.3 24.2 7.2
Neither agree
nor disagree 0.0 12.0 21.4 15.2 13.0
Agree 100.0 76.0 53.6 57.6 63.3
Strongly agree 0.0 0.0 10.7 3.0 13.5
* Correlation is signicant at the 0.05 level.
78
SOCIAL MEDIA & POLITICAL CAMPAIGNS IN KENYA | John O. Ndavula
79
Chapter 3 Diffusion of Social Media in the 2013 Election Campaigns
Table 3.12: Literacy Levels of Constituents and Social Media Adoption
Electoral Ofce (%) Chi-square
test
President
(n =6)
Governor
(n=25)
Senator
(n=28)
Women Rep.
(n = 34)
MP
(n=207)
Disagree 0.0 0.0 10.7 5.9 4.8
X2 = 6.661
df = 12
Sig = 0.879*
Neither agree
nor disagree 16.7 4.0 7.1 14.7 7.7
Agree 16.7 60.0 60.7 55.9 58.9
Strongly agree 66.7 36.0 21.4 23.5 28.5
* Correlation is signicant at the 0.05 level.
Table 3.13: Model Specication and Denition of Variables
Dependent variable
Use of social media: “Did you use any social media for your campaign in the 2013
General Elections?” (Dummy variables, 1 for Yes and 0 for No)
Independent variables
1. Familiarity with technology measured as whether extremely influential, very
influential, somewhat influential, slightly influential, not at all influential.
2. Familiarity with technology measured as whether extremely influential, very
influential, somewhat influential, slightly influential, and not at all influential.
3. Level of ofce measured as whether extremely influential, very influential,
somewhat influential, slightly influential, and not at all influential.
4. Level of competition measured as whether extremely influential, very influential,
somewhat influential, slightly influential, and not at all influential.
5. Established tradition measured as whether extremely influential, very influential,
somewhat influential, slightly influential, and not at all influential.
6. Financial consideration measured as whether extremely influential, very influen-
tial, somewhat influential, slightly influential, and not at all influential.
7. Political party influence measured as whether extremely influential, very influen-
tial, somewhat influential, slightly influential, and not at all influential.
8. Gender measured as either male or female.
9. Age measured as below 20 years, 21–35 years, 36–50 years, 51 years and above.
10. Education measured as diploma, degree, masters’, doctorate.
11. Title of ofce measured as president, governor, senator, women representative,
Member of Parliament.
12. Party afliation measured as The National Alliance, Orange Democratic Party, and
United Democratic Forum.
13. Awareness of use of social media for election campaigns measured as yes or no.
14. Years of internet use measured as more than 10 years, 3–10 years, less than 3
years, and no experience.
15. Type of constituency measured as urban, peri-urban, and rural.
80
SOCIAL MEDIA & POLITICAL CAMPAIGNS IN KENYA | John O. Ndavula
81
Chapter 3 Diffusion of Social Media in the 2013 Election Campaigns
Table 3.14: Correlations of the Model Variables
Constant Familiarity Ofce Competition Tradition Financial
Constant 1.000 -.023 .101 -.096 -.101 .011
Familiarity with
technology -.023 1.000 -.573 .006 .269 .184
Level of ofce .101 -.573 1.000 -.268 -.175 -.162
Competition
level -.096 .006 -.268 1.000 .044 -.182
Tradition -.101 .269 -.175 .044 1.000 -.326
Financial .011 .184 -.162 -.182 -.326 1.000
Party influence -.454 -.413 .174 -.019 -.114 -.380
Gender -.370 .060 .168 -.021 .064 -.085
Age -.526 .266 -.131 .103 .400 -.227
Education -.028 .117 -.287 .221 .082 -.002
Level of ofce -.454 .103 -.332 -.035 .193 .126
Party afliation -.449 -.390 .274 -.123 -.303 -.198
Social media
use -.316 -.474 .132 -.040 -.381 -.122
Years of internet
use -.338 .143 -.166 .095 .096 -.074
Constituency
type .009 .032 -.007 -.060 -.214 .262
Party
Influence Gender Age Education Level of
Ofce Party
Social
Media
Use
Years of
Internet Constituency
-.454 -.370 -.526 -.028 -.454 -.449 -.316 -.338 .009
-.413 .060 .266 .117 .103 -.390 -.474 .143 .032
.174 .168 -.131 -.287 -.332 .274 .132 -.166 -.007
-.019 -.021 .103 .221 -.035 -.123 -.040 .095 -.060
-.114 .064 .400 .082 .193 -.303 -.381 .096 -.214
-.380 -.085 -.227 -.002 .126 -.198 -.122 -.074 .262
1.000 .161 .174 -.193 .085 .578 .388 .061 -.115
.161 1.000 .364 -.228 -.074 .176 -.213 .231 -.052
.174 .364 1.000 -.094 .003 .192 -.130 .364 -.462
-.193 -.228 -.094 1.000 .232 -.245 -.173 -.159 .097
.085 -.074 .003 .232 1.000 -.004 -.094 -.227 .104
.578 .176 .192 -.245 -.004 1.000 .422 .168 -.320
.388 -.213 -.130 -.173 -.094 .422 1.000 .020 .026
.061 .231 .364 -.159 -.227 .168 .020 1.000 -.273
-.115 -.052 -.462 .097 .104 -.320 .026 -.273 1.000
82
SOCIAL MEDIA & POLITICAL CAMPAIGNS IN KENYA | John O. Ndavula
83
Chapter 3 Diffusion of Social Media in the 2013 Election Campaigns
Table 3.15: Binary Logistical Regression Statistics
Omnibus Tests of Model Coefcients
Chi-square df Sig.
Step 1
Step 80.639 14 .000
Block 80.639 14 .000
Model 80.639 14 .000
Model Summary
Step 1 -2 Log likelihood Cox & Snell R Square Nagelkerke R Square
91.199a.273 .754
a. Estimation terminated at iteration number 8 because parameter estimates
changed by less than .001.
Classication Tablea
Observed
Predicted
Did you use any media for your
campaign in 2013 general election? Percentage
Correct
No Yes
Step 1
Did you use any
media for your
campaign in 2013
general election?
No 10 17 37.0
Yes 3 223 98.7
Overall Percentage 92.1
a. The cut value is .500
Variables in the Equation
B S.E. Wald df Sig. Exp(B)
Step 1a
Familiarity 1.527 .425 12.907 1 .000* 4.602
Ofce -.422 .428 .975 1 .323 .655
Competition .412 .274 2.265 1 .132 1.509
Tradition .454 .291 2.439 1 .118 1.575
Financial .071 .312 .051 1 .821 1.073
Party influence -.600 .285 4.421 1 .036* .549
Gender .352 .754 .218 1 .641 1.422
Age .233 .508 .211 1 .646 1.263
Education .603 .450 1.798 1 .180 1.828
Ofce
campaigned -.243 .427 .324 1 .569 .784
Party aflia-
tion -.681 .301 5.122 1 .024* .506
Social media
awareness 6.922 1.943 12.684 1 .000* .001
Years of
internet .521 .421 1.529 1 .016* 1.683
Constituency
rating .057 .443 .017 1 .897 1.059
Constant 4.693 3.965 1.401 1 .237 109.188
a Variable(s) entered on step 1: familiarity, ofce, competition, tradition, nancial, party influence,
gender, age, education, ofce campaigned, party, social media awareness, years of internet, and
constituency rating.
* Correlation is signicant at the 0.05 level
85
CHAPTER 4
Techniques for Campaigning on Social Media
One social media site manager I interviewed said: “I think it is not
that dicult to use social media… It doesn’t really need training per se
so long as you can follow instructions.” is statement exemplies the
attitude that most candidates have towards social media. However,
candidates need to be aware of techniques of campaigning on social
media, since campaigns’ practices are inscribed features that those
social media possess. It is for this very reason that social networks re-
quire technical expertise and specialised knowledge to leverage them
eectively.1 In this chapter, I map out the techniques that candidates
use to campaign on social media. I assess the prevalence of techniques
of informing, involving, connecting, and mobilizing. Informing in-
volves presenting information to site visitors. Involving requires the
distribution of information about a campaign to recruit supporters to
participate. Connecting entails enabling an interaction or exchange of
information between site visitors and other political actors. Mobilizing
involves equipping supporters to promote the candidate to others, that
is, to become information disseminators on behalf of the campaign.
To ascertain how these techniques were manifested on social media, I
1. Williams and Gulati, “Social Networks in Political Campaigns,” 54.
86
SOCIAL MEDIA & POLITICAL CAMPAIGNS IN KENYA | John O. Ndavula
87
Chapter 4 Techniques for Campaigning on Social Media
conducted in-depth interviews with candidates and social media site
managers from various campaigns. Using the data obtained, I assess
how these practices were adopted across the social media platforms for
election campaigns and put forward some specic suggestions on how
to approach election campaigns on social media. I conclude the chap-
ter by situating social media campaigns within the broader context of
political marketing.
Techniques of Campaigning on Social Media
ere are four techniques of campaigning on social media that are
particularly signicant to electoral campaigning.2 ese techniques
are informing, involving, connecting, and mobilizing. Informing in-
volves presenting information to site visitors. Involving requires the
distribution of information about a campaign to recruit supporters to
participate. Connecting entails enabling an interaction or exchange
of information between site visitors and other political actors. Lastly,
mobilizing involves equipping supporters to promote the candidate to
others, that is, to become information disseminators on behalf of the
campaign. ese techniques allow for informing prospective voters,
involving supporters, connecting users with other political actors,
and mobilizing supporters to become advocates, and they are utilized
by campaigns in diering ways and to varying degrees. We shall now
focus our attention on each of these techniques in turn.
TECHNIQUES OF INFORMING
Informing is foundational for all online structures that aim to
present information to site visitors.3 Generally, campaigns have a man-
date to create and maintain a regular supply of current information.
Candidates can share documents on social media, including policy
statements and their positions on specic issues. Candidates can also
2. Foot and Schneider, Web Campaigning, 47.
3. Ibid., 47.
post biographies, issue statements, texts of speeches delivered by the
candidate, press releases and other campaign news, and text or video
presentations of campaign advertisements released in other media.4
Candidates can choose to highlight issues on which they have a
stand that is likely to appeal to the most voters. Candidates can also
integrate online and oine realms of production. ey can post mate-
rial from non-digital media such as print, lm, broadcast radio, and
television. Texts originally created on analog or oine media, such as
typewritten and photocopied yers, can be digitized. In addition, can-
didates can post video or audio les that were originally produced for
dissemination on television or radio.5 is can be considered a form of
technological convergence.
Candidates may post links to material posted on a dierent site. is
can provide more extensive information, while keeping the campaign
site distinct from those other sites. A campaign’s choices regarding the
links provided can shape users’ perceptions of the informational value
of the site. is can facilitate expansiveness in the breadth and depth of
the materials, while enabling site visitors to select the type and level of
information they wish to view. For example, a candidate can provide a
link to a site that criticizes their opponents.
Dierent branches of a campaign can jointly produce content for
a social media site. A variant of this is when a candidate reproduces
material previously distributed by an external organization without
providing a link to that organization. For example, a campaign can
present scanned images of newspaper articles on their sites. ese
images in turn represent the candidate’s presence in the oine news
media and convey the message that the candidate’s views and activities
are “newsworthy” and thus should be attended to by site visitors.6
4. Ibid., 47.
5. Ibid., 56.
6. Ibid., 60.
88
SOCIAL MEDIA & POLITICAL CAMPAIGNS IN KENYA | John O. Ndavula
89
Chapter 4 Techniques for Campaigning on Social Media
TECHNIQUES OF INVOLVING
Campaigns engage in the web practice of involving when they cre-
ate and provide online structures facilitating interaction between site
visitors and the campaign organization. e simplest form of involving
is providing information that can allow a site visitor to contact the
campaign team, such as a telephone number or email address. A visitor
can indicate willingness to volunteer time or eort or make a nancial
contribution. Users can also be invited to like, retweet, or favourite a
message. Involving as a campaign practice is a familiar and traditional
part of electioneering.7
Transactional relationships that require site visitors to provide
information can also be explored. is information can be used for
processing donations and managing volunteer relationships oine.
However, online visitors may be more sensitive to and have more
concerns about engaging in online transactions with campaigns than
if the same request were to be made oine.8 e campaign site can also
provide documents which can be downloaded and distributed, such as
posters and yers.
TECHNIQUES OF CONNECTING
e third online campaigning practice is connecting. e aim of
connecting is to facilitate site visitors’ interaction with other political
actors. A campaign may associate itself with other political actors by
directly referencing others in the content of texts or images on its
social media site. For instance, association may be manifested through
explicit statements of aliation with a political party.9 A campaign can
create a structure that encourages a user to engage with another actor
in the oine world. Visitors can also contribute content, such as mes-
7. Ibid., 70.
8. Ibid., 77–78
9. Ibid., 105.
sages, and hence enable contributors to connect with each other.10 e
candidate creates links from their site to help create an electoral web
sphere, within which citizens and others who use the web can engage
with each other in various political actions.
TECHNIQUES OF MOBILIZING
Mobilizing aims at building a community of advocates for the
candidate. Mobilizing is the act of using social media to persuade and
equip campaign supporters to promote the candidate to others, both
online and oine. is technique can include provision of electronic
versions of brochures, yers, or other documents promoting the candi-
date, along with encouragement to visitors to download and distribute
these documents. Campaigns can use their sites to equip supporters
to produce their own material for distribution in oine media. ey
can provide supporters with tips, suggested campaign messages, or
talking points for supporters to personalize. By allowing themselves to
be mobilized via the social media to engage in candidate promotion,
supporters participate in coproducing the campaign along with the
candidate and campaign sta.11 Convergence can also be manifested
by using social media to mobilize supporters to go out and vote.
Site visitors can engage in coproduction through the display and
dissemination of an individual’s endorsement or message of support
to members of the endorser’s social network. is can also be achieved
by encouraging interlinking of the campaign’s social media site with
supporters’ sites.12
Empowerment in the context of mobilizing can be understood as
the provision of online structures that especially facilitate visitors to
engage in political actions that are seemingly beyond the control of
the campaign and without the campaign’s knowledge. Web visitors can
create parodies or distribute content for the candidate. Candidates can
10. Ibid., 112.
11. Ibid., 131.
12. Ibid., 146–147.
90
SOCIAL MEDIA & POLITICAL CAMPAIGNS IN KENYA | John O. Ndavula
91
Chapter 4 Techniques for Campaigning on Social Media
write letters to editors, call into talk shows, host a party, and so on. Most
of the features used for mobilizing are like those used for connecting.13
Techniques Employed in the 2013 Election Campaigns
Informing was the major technique employed in the 2013 election
campaigns in Kenya. Informing involves presenting information to site
visitors. Candidates created and maintained a steady supply of infor-
mation for voters to consume. ey posted text messages, photos, and
videos to varying degrees (see Figure 4.1). Findings indicate that most
of the candidates sent text messages. Other candidates posted images
and video messages.
13. Ibid., 136.
A social media site sta member for a presidential campaign candi-
date interviewed stated:
“Our campaign was structured in such a way that as we moved
towards the election, there were dierent things we were high-
lighting; from the launches, to the manifestos, to the rallies. If
we were doing a launch, we would post the entire speech, if
presenting a manifesto, we would post the entire manifesto, if
we were conducting a rally, we would post updates about the
rally as it went on. In addition, we would update voters on our
campaign trail. We would tell them that Jubilee team will be in
Bungoma at this time, at this time in Kandui, and at this time
Webuye. Our updates would include both text messages and
pictures to make them follow us constantly.”
Although text was the most popular message form, some social me-
dia site managers challenged the use of text, indicating a preference for
images. A social media strategist for a gubernatorial candidate stated:
“I would say photos are more important than text because they
speak much. Text may fail to go viral, but photos can. We have
seen photos going viral of development [issues]. Also, a single
photo with a small caption—for example ‘this is where I met
several people’—adds a lot of value. Photos are worth a thou-
sand words. ere is evidence that you were at a location. For
example, if you take a photo at a local market or a bodaboda
stop, people can be able to identify with it. It can establish an
emotional connection.”
In general, most campaigns were geared towards supplying infor-
mation to voters (see Figure 4.2). e information given included
general communication with site visitors, sharing party and candidate
positions, disclosing achievements, appealing to voters, countering
propaganda by sharing a position on a policy issue, and disclosing
qualications. Interviews with candidates also reveal that much of their
eort on social media was geared towards defense against propaganda.
One senatorial candidate observed that:
“If there is propaganda going on via social media, it can
negatively aect the politician’s campaign and hence inuence
the election outcome negatively. is is more so if you have
Figure 4.1: Nature of Messages Sent
92
SOCIAL MEDIA & POLITICAL CAMPAIGNS IN KENYA | John O. Ndavula
93
Chapter 4 Techniques for Campaigning on Social Media
opinion leaders who act as inuencers. ey could go to their
networks and state a particular negative view and say, ‘Even
on social media so many people are saying this is true.’”
is implies that a signicant number of candidates used social
media for propaganda purposes. According to social marketing theory,
one of the most potent promotional strategies in electoral campaigns
is staging negative campaigns, that is, staging a direct or indirect com-
parative assault against the position of the opponent and their personal
characteristics.14 e fact that candidates were responding to negative
campaign messages demonstrates that a signicant number of political
candidates in Kenya employ propaganda or negative campaigns on
social media as a marketing strategy.
14. Iyengar, Media Politics, 173–174.
e second technique that campaigns utilized to engage with voters
on social media is mobilizing (see Figure 4.3). Mobilizing is using so-
cial media to persuade and equip campaign supporters to promote the
candidate to others, both online and oine. Findings indicate most of
the candidates used social media to mobilise supporters to attend ral-
lies and demonstrations. Some used social media to recruit volunteers,
while a few others solicited for funds. Although in some developing
countries, social media have been used successfully for fundraising,
in the Kenyan context, fundraising ranks as the least-used strategy. In
Kenya, candidates are more inclined to issue cash handouts to voters
during their campaigns to inuence their voting decision. e practice
of candidates raising campaign funds from voters is not institutional-
ized in Kenya, hence demarcating politics as an arena for candidates
who are already nancially endowed. Furthermore, there are no clear
means of channelling cash raised to campaigns, although mobile
money transfer systems like M-Pesa through the pay bill facility may
soon ll this void, if adopted. Lack of a legal support framework for
fundraising also impedes online fundraising for campaigns.
Figure 4.2: Techniques of Informing Voters
Figure 4.3: Techniques of Involving and Mobilizing Voters
94
SOCIAL MEDIA & POLITICAL CAMPAIGNS IN KENYA | John O. Ndavula
95
Chapter 4 Techniques for Campaigning on Social Media
e third technique used by candidates on social media was involv-
ing. Involving requires the distribution of information about a cam-
paign to recruit supporters to participate. Most of the candidates used
social media to announce events to their site visitors with the aim of
getting them involved in the campaign events (see Figure 4.3).
None of the candidates surveyed employed connecting as a strategy
for campaigning on social media. Connecting entails enabling an in-
teraction or exchange of information between site visitors and other
political actors. It is uncertain at this point whether this is a strategy
that campaigns will adopt in the future, or whether there are any factors
intrinsic to the Kenyan political scene that make this strategy unlikely
to be adopted.
Discussion on Techniques of Using Social Media
All campaigns with an online presence engaged in the practice of
informing. Informing seems to be a relatively straightforward prac-
tice. However, candidates and site producers must contend with the
accuracy, volume, and currency of the information they produce. Pro-
ducing accurate content, and keeping it accurate during an evolving
campaign, requires a lot of eort. It is beyond the scope of this book
to evaluate the quality or value of the information provided by cam-
paigns on social media, or to compare it to campaign information in
other media. However, social media oer opportunities for expanded
breadth and depth of information at relatively low cost and that many
campaigns are taking advantage of that opportunity.
Some campaigns utilized the technique of mobilizing. However,
campaigns that engage in mobilizing walk a tightrope because it brings
in issues of control. Candidates cannot force social media site visitors
to become advocates for their campaigns, whether online or oine.
Nor can they force any site visitor to use the material they post online
in the ways the campaign intends that material to be used, such as
sharing with others. is is a big challenge because ultimately social
media site visitors have the nal say on what actions they take online.
Few campaigns were shown to have utilized the technique of involv-
ing yet involving is instrumental in providing contact information and
establishing transactional relationships with site visitors. In general,
more complex forms of involving, in which campaign organizations
obtain information or data provided explicitly and consciously by site
visitors, are not common in current Kenyan political campaigns. e
lower incidence of these facets of involving can be explained in part by
their distinctiveness by the necessity for campaign organizations to es-
tablish a transactional relationship with site visitors. is requirement
dictates a distinct set of skills and a commitment of resources at a dif-
ferent scale than those necessitated by the practice of informing. Most
generally, adaptation of this practice requires campaign organizations
to be able to request, accept, and process data provided by site visitors, to
create and manage a transactional relationship with site visitors. ere
are two implications of this requirement. First, the technical expertise
demanded by the practice of involving is considerably dierent than
that required to simply produce and distribute information on social
media. Second, the practice of involving necessitates an entanglement
of the campaign organization with site visitors far beyond the familiar
push or broadcast pattern found with the practice of informing.15
ese factors are exhibited in the tensions associated with the practice
of involving, and they explain in part why these practices are not more
widespread.
None of the candidates used connecting techniques on social media.
is can be seen as a missed opportunity, as the practice of connecting
helps in shaping candidate identity, establishing candidate credibility,
building community, and extending resources. In general, fewer cam-
paign organizations engage extensively in the practice of connecting.16
A reason for this failure could be that candidates fear losing control
over the message of the campaign. Over time, the other political actors
may become less desirable associations for the campaign. Furthermore,
15. Foot and Schneider, Web Campaigning, 99.
16. Ibid., 123.
96
SOCIAL MEDIA & POLITICAL CAMPAIGNS IN KENYA | John O. Ndavula
97
Chapter 4 Techniques for Campaigning on Social Media
linking to another actor exposes site visitors to the messages produced
by that other actor. When messages produced by other actors become
negative, the value of establishing the connection diminishes. In ad-
dition, campaigns risk losing the attention of visitors to the other site
they have connected to. Overall, campaigns are inherently risk-averse
organizations and therefore they may perceive the practice of connect-
ing as risky.17
Social Media and Political Marketing
Social media provide political candidates with unlimited space
and innovative ways of articulating their campaigns.18 Political can-
didates can maximise the potential of using social media for election
campaigns by incorporating political marketing approaches in their
campaign design. Political marketing is about political organisations
adapting business marketing concepts and techniques to help them
achieve their goals. Political marketing derives from social marketing
theory. is theory posits that marketing principles, tools, and tech-
niques can be applied to create, communicate and deliver information
to inuence the target audience.19 Social marketing packages a product
and utilises the optimum combination of campaign components to
attain pragmatic goals. e specic products that are marketed are the
political candidates, political parties, and the benets voters believe
will result if the candidate is elected.20 e candidates have an oppor-
tunity to spell out their major voting promises on their social media
sites. Whereas the goal of marketing is to attract new customers and to
17. Ibid., 127.
18. John O. Ndavula and Hellen K. Mberia, “Social Networking Sites in Kenya:
Trigger for Non-institutionalized Democratic Participation,” International Journal of
Business and Social Science 3:13 (2012): 300.; Tedesco, Miller and Spiker, “Presidential
Campaigning on the Information Superhighway,” 51–63.
19. Philip Kotler, Ned Roberto and Nancy Lee, Social Marketing: Improving the
Quality of Life (ousand Oaks, CA: Sage, 2002), 4–20.
20. Jennifer Lees-Marshment, “e Marriage of Politics and Marketing,” Political
Studies 49 (2001): 692–713.
keep and grow current customers, the goal in election campaigns is to
maintain current voters and to win over new ones.
For online election campaigns to be eective, candidates and site
managers need to adopt appropriate social media techniques such as
informing, involving, connecting, and mobilizing to market themselves
eectively. For instance, social networking sites can be used eectively
for the recruitment of volunteers, organization of the campaign, mobi-
lization, and fundraising.21
Candidates could choose to deploy the long-tail marketing strategy
for campaigns on social media. Long-tail marketing involves candi-
dates spreading their eorts across several dierent channels, each
with specic intended target groups and tailored messages.22 Due to
the vast amount of user info gathered by social media sites, they are
well suited for sending tailored campaign messages to specic voter
groups. e candidate can leverage various sites such as Facebook and
Twitter to engage audiences who have an anity for those media.
Fourth, social media hold a potential to bypass the requirement of
active user choices for them to become exposed to campaign messages.
Previously, political websites tended to mostly attract citizens with a
strong interest in politics and not those who stumble across political
content accidentally.23 Candidates employing social networking sites
21. Derrick Cogburn and Fatima Espinoza-Vasquez, “From Networked Nominee
to Networked Nation: Examining the Impact of Web 2.0 and Social Media on Politi-
cal Participation and Civic Engagement in the 2008 Obama Campaign,” Journal of
Political Marketing 10: (2011): 189-213.; Kay L. Schlozman, Sidney Verba, and Henry
E. Brady, “Weapon of the Strong? Participatory Inequality and the Internet,” Perspec-
tives on Politics, 8:2 (2010): 487–509.; Greyes, “e Untapped Potential of Social
Media,” 45–47.; Will Straw, “Yes We Did? What Labour Learned from Obama,” in
e internet and the 2010 Election: Putting the Small ‘p’ Back in Politics?, eds. Rachel
Gibson, Andy Williamson & Stephen Ward (London: Hansard Society, 2010), 43–50.;
Maria L. Sudulich and Mathew Wall, ‘Every Little Helps’: Cyber-Campaigning in the
2007 Irish General Election,” Journal of Information Technology & Politics, 7 (2010):
340–355.
22. Maria L. Sudulich et al., “Me Too for Web 2.0?”
23. Pippa Norris, Digital Divide? Civic Engagement, Information Poverty and the
Internet Worldwide (New York: Cambridge University Press, 2001b), 118–195.
98
SOCIAL MEDIA & POLITICAL CAMPAIGNS IN KENYA | John O. Ndavula
99
Chapter 4 Techniques for Campaigning on Social Media
in their campaigns, however, have a potentially higher probability of
reaching citizens beyond those who visit their social networking sites.
is ability to potentially start a viral “chain reaction” that touches a
broader online audience corresponds to the ways in which campaigning
through traditional mass media was able to reach passive “viewers.”24
Cogburn and Espinoza-Vasques credit part of the success of the
Obama campaign in the 2008 US elections to online visitors spread-
ing the campaign messages onwards.25 Another “indirect” impact of
social media campaigning is the increasing habit of media journalists
to use social networking sites as a news source, which gives messages
originating online augmented publicity online and also oine.26
e application of the marketing concept in politics may result in
politics becoming more democratic. Political marketing can improve
the quantity and quality of information ows from the electorate to
parties and candidates, thus making them more sensitive and respon-
sive to voters’ needs. At the same time, it improves the channels of
communication from candidates to the electorate and even more to
every specic segment of voters. en, “‘political marketing’ provides
a rational way for parties or candidates to behave in conditions of
competitive mass democracy”.27
Conclusion
In this chapter I have described the techniques for campaigning using
social media. e techniques considered include informing, involving,
connecting, and mobilizing. Informing involves presenting informa-
24. Greyes, “e Untapped Potential of Social Media,” 45–47.
25. Cogburn and Espinoza-Vasquez, “From Networked Nominee to Networked
Nation,” 189–213.
26. Nils Gustafsson, “e Subtle Nature of Facebook Politics: Swedish Social
Network Site Users and Political Participation,” New Media & Society 14:7 (2012):
1111–1127.
27. Margaret Scammell, Designer Politics: How Elections are Won (Basingstoke:
Macmillan Press, 1995), 18–19.
tion to site visitors. Involving requires the distribution of information
about a campaign to recruit supporters to participate. Connecting
entails enabling an interaction or exchange of information between
site visitors and other political actors. Mobilizing involves equipping
supporters to promote the candidate to others, thus becoming infor-
mation disseminators on behalf of the campaign. Data presented in
the chapter indicates that informing is the most prevalent social media
campaign strategy. A few other candidates used to involve and mobi-
lizing. However, none of the campaigns used connecting techniques.
Although the early history of politics on the internet suggested the po-
tential for alternative modes of campaigning, informing emerged as the
initial and dominant practice of campaigning in Kenya. is chapter
also explored the potential of adopting social marketing principles to
structure election campaigns on social media. Network society theory
states that political actors have the capacity to act on social networks,
a process which results in reconguration of the network according
to their needs, desires, and projects.28 Indeed, candidates in Kenya
utilise social media in unique ways by adapting the media to their own
particular campaigning needs.
28. Manuel Castells, “Informationalism, Networks, and the Network Society: A
eoretical Blueprint, in e Network Society: A Cross-cultural Perspective, ed. M.
Castells (Northampton, MA: Edward Elgar Publishing, 2004), 3–45.
101
CHAPTER 5
The Political Impact of Social Media
Considerable research devoted to the question of how voters arrive
at their decisions to elect certain candidates indicates that there are no
easy answers to this question. Some of the notable factors that inu-
ence voters’ choices are political party, policy preferences, candidate,
ethnicity, gender, and education.1 Whatever the combination of factors
that come into play in any election, certainly the media in which these
aspects are communicated need closer scrutiny. Voters, who ultimately
decide the outcome of any election, rely on the media for campaign
information. In this chapter I explore the eectiveness of social media
as channels for electoral campaigns in Kenya. I also consider the pros-
pects and challenges of using social media to reach the younger age
demographic. Finally, I situate social media under the larger discourse
of electronic democracy. I attempt to nd a middle ground between
optimists who see social media as progressive and improving the qual-
ity of election campaigns and pessimists who regard the claims made
for social media as overblown and unrealistic.
1. William L. Benoit, Communication in Political Campaigns (New York: Peter
Lang, 2007), 213.
102
SOCIAL MEDIA & POLITICAL CAMPAIGNS IN KENYA | John O. Ndavula
103
Chapter 5 The Political Impact of Social Media
Social Media and Election Outcome
All political candidates are concerned with the question of which
medium is most eective in reaching voters with campaign messages.
e focus of this section is on the eectiveness of social media as chan-
nels for electoral campaigns. Everett Rogers posits that innovations
whose results are observable are adopted more quickly than other
innovations.2 In this sense, then, the observability of social media lies
in their ability to impact election outcomes. If using social media can
potentially inuence voting decisions, then it is likely that the popular-
ity of the media among political candidates will continue to grow in
subsequent elections.
I sought to nd out political candidates’ view of the impact of social
media on political outcome in Kenya. Data indicates that candidates
who participated in the 2013 elections expressed ambivalence on the
contribution of social media towards success in Kenyan elections (see
Table 5.1). Further analysis indicates that the popularity of a candidate
on social media was not an indicator of actual popularity on the ballot.
Candidates at the presidential level who had the most likes on Face-
book, except for Uhuru Kenyatta, failed to register similar popularity
on the ballot (see Figure 5.1). For example, Raila Odinga and Musalia
Mudavadi emerged the rst and second runners up, yet their Facebook
likes were signicantly lower than their rivals Martha Karua and Peter
Kenneth.
A similar trend was witnessed from the analysis of presidential
candidates on Twitter. e most followed candidates on Twitter were
not necessarily the candidates who received the most votes (see Figure
5.2). Liking and following on social media was not an indicator of
candidate popularity.3
2. Rogers, Diusion of Innovations, 16.
3. John O. Ndavula and Joy Mueni, “New Media and Political Marketing in Kenya:
e Case of 2013 General Elections,” International Journal of Arts and Commerce 3:6
(2014): 69–84.
Figure 5.1: Facebook Likes and Actual Votes
Figure 5.1: Twitter Followers and Actual Votes
104
SOCIAL MEDIA & POLITICAL CAMPAIGNS IN KENYA | John O. Ndavula
105
Chapter 5 The Political Impact of Social Media
e interviews I conducted with candidates and social media site
managers further reinforced their ambivalence about the adoption of
social media. Some of the candidates expressed optimism about social
media’s contribution to election outcome while others did not. A par-
liamentary candidate speaking in favour of social media use stated:
“Social media can aect election outcome by a small margin.
It largely depends on the context of the election, whether it is
national, urban or rural. If it is in urban areas where uptake
of social media is [sic] higher, use of social media will have a
wider audience. In rural areas, the reach is minimal because
of lack of access. But social media have the power and capacity
to do that.”
On the other hand, some candidates were less optimistic. A social
media site manager for a gubernatorial candidate stated:
“A politician who did not use social media can get more votes
than the one who used it because there is a phenomenon we
call ‘voting on social media.’ You can be the darling of Ken-
yan social media users but that doesn’t necessarily translate
into votes.”
ere are several reasons that may account for a diminished impact
of social media on voting decisions. First, Kenya is polarized along eth-
nic lines and hence has witnessed ethnic voting blocs famously referred
to as “the tyranny of numbers” in previous elections. Second, followers
of a candidate on social media may not necessarily be registered voters
or even drawn from their constituency. It is also possible that some live
in the diaspora and may not be available to cast their ballot.
Despite this, it is surprising to note that most political candidates
interviewed expressed optimism about the success of social media for
election campaigns in Western democracies, especially in US elections
(see Table 5.2). If we turn our attention to the US elections, we note
that the inuence of online campaigns in general and social media on
election outcome has been growing. In the initial phases of internet
adoption in the US, social media was not very instrumental in deliv-
ering votes, but it increasingly became so. e rst campaign to take
advantage of the internet to organize its supporters was Jesse Ventura’s
1998 campaign for governor of Minnesota. Ventura was heavily out-
spent by his competitors, but he launched a campaign website with
one paid sta, the campaign manager. Although Ventura was elected
governor, it is dicult to say precisely how much dierence Ventura’s
use of technology made to the upset victory. In one respect, however,
the payo of using technology was clear: young people, who are most
likely to be reached online, turned out to vote in large numbers. Over
half of them voted for Ventura—more than enough to account for the
margin of his victory.4
e success of the Ventura campaign made it clear to campaign
operatives that the internet could and should be exploited for political
action. John McCain used the internet extensively in the 2000 presi-
dential primary. Despite McCain’s initial success, he lost the race.5
In the 2004 elections, Governor Howard Dean’s innovative use of
technology contributed to his meteoric rise from the relatively un-
known governor of a small state to the front-running contender for the
Democratic nomination. Dean’s most accomplished feat was his ability
to raise funds online. However, Dean’s extensive network of online
volunteers did not translate into an outpouring of votes, and he was
forced to withdraw from the race.6 us, the main lesson of 2004 was
that although the internet provided a cost-ecient means of develop-
ing a network of campaign workers and donors, it was not yet the best
platform for candidates to appeal for votes.7 In other words, it is one
thing to develop an electronic network of enthusiastic supporters, and
it is quite another to attract enough votes to win an election.
4. Iyengar, Media Politics, 136–139.
5. Ibid., 137–139.
6. Allison Slotnick, “‘Friend’ the President: Facebook and the 2008 Presidential
Election,” in Politicking Online: e Transformation of Election Campaign Communi-
cations ed. Costas Pangopoulos (New Jersey: Rutgers, 2009), 250.
7. Iyengar, Media Politics, 138.
106
SOCIAL MEDIA & POLITICAL CAMPAIGNS IN KENYA | John O. Ndavula
107
Chapter 5 The Political Impact of Social Media
In 2008, Barack Obama’s presidential campaign demonstrated that
it was possible for the internet platform to be harnessed not just to
mobilize activists and raise money, but also to win votes. He used all
the major social networking sites, like Facebook and Twitter, and went
on to win every single caucus that mattered.8 Gulati and Williams con-
tend that evidence from their various models and analyses makes the
case that Facebook played a role in both the 2006 congressional races
and early 2008 nomination contests. ey oer some initial empirical
conrmation that social networking sites indeed have the potential to
transform campaigns and the electoral process. eir study demon-
strates empirically that the use and success of online campaigning must
be considered side-by-side with traditional methods.9 In addition, the
study calls attention to the use of social media to reach the younger age
demographic.
In the 2016 US elections, social media provided voters with insights
into the thinking of the political candidates in a way that was never
possible. For example, to see what went on inside the mind of Re-
publican presidential candidate Donald Trump, one could study his
personal Twitter account, which had 12.5 million followers. Trump
used social media to criticize opponents in an unprecedented negative
campaign. Democratic candidate Hillary Clinton, who had 9.8 million
Twitter followers, used social media to hit back at Trump’s criticisms.
Some candidates show their true selves on social media, at the risk of
enraging or oending voters. Others hide behind professional, bland
accounts, risking being perceived as characterless. Overall, candidates
are trying to nd new ways to reach people who are increasingly con-
suming news through online platforms.10
8. Ibid., 139.
9. Christine B. Williams and Girish J. Gulati, “e Political Impact of Facebook:
Evidence from the 2006 Midterm Elections and 2008 Nomination Contest,” Politics
and Technology Review 1:1 (2008): 11–24.
10. Jim Pickard, “When Politics and Social Media Collide.” Accessed November
22, 2016. https://www..com/content/27a7d6c8-702f-11e6-a0c9-1365ce54b926
Unless these ndings of social media success are specic to the US
context, we expect candidates who campaign on social media to have a
relative vote advantage over those who do not, when controlling for all
other variables. Yet, caution should be taken when making judgments
about the actual contribution of supporters on social media to the can-
didates’ ultimate margin of victory, given that 18–29 year olds have a
lower voter turnout rate than other age groups. With individuals under
18 comprising 14 percent of the community on social media gener-
ally, there is a sizable group of site visitors who are not even eligible to
vote.11 As noted earlier, social media site visitors could express support
for multiple candidates or live outside the candidates’ constituencies or
counties. Other reasons for exercising caution are that some members
of Facebook and Twitter may not be intending to vote or even regis-
tered to vote; therefore, it is not possible to actualize their support for
candidates on social media.
Reaching the Youth
Direct candidate-to-voter communication through social media is
likely to have the greatest impact on younger Kenyans because they
compose most computer-based media users. e youth are generally
adept with new technologies and have integrated them into their per-
sonal lives as never before. e important question is how candidates
can use social media eectively to reach the technology generation.
e youth are attracted to technology not because they seek political
enlightenment but rather for social interaction and personal gratica-
tion. If technology-enhanced political material is to catch their eye,
the presentation must necessarily include popular elements of youth
culture, most notably music, animation, and comic relief.12
Conventional election campaigns provide the youth with an external
rationale for voting. Yet external motivation may impede the develop-
11. Iyengar, Media Politics, 135–146.
12. Ibid., 342–343.
108
SOCIAL MEDIA & POLITICAL CAMPAIGNS IN KENYA | John O. Ndavula
109
Chapter 5 The Political Impact of Social Media
ment of participant attitudes. e fact that the youth, or any individuals
for that matter, are free to explore social media sites on their own has
important psychological implications. Typically, individuals attribute
their own actions to either internal or external causes. For instance, a
youth who votes may believe that they decided to vote on their own or,
alternatively, that they were inuenced to vote. Attributing the act to
internal factors contributes to intrinsic motivation, which is known to
encourage long-term learning of the act in question.13
e implications for intrinsic motivation are that young people who
seek out and encounter campaign information of their own accord
and spend time interacting with political material may come to see
themselves as interested in politics. e relatively inexpensive act of
visiting a social media site may then open the door to more signicant
acts, including registering to vote and discussing the campaign with
friends or parents. us, a relatively trivial and unobtrusive addition
to one’s technology space promises greater long-term payos than do
conventional eorts at mobilization.14
erefore, campaigns can reap more benets online by engaging
in the practices of informing, mobilizing, involving, and connecting
when targeting the youth (see chapter 4). Site producers must ensure
accuracy, volume, and currency of the information they produce. ey
also need to build a community of advocates for the candidate running
for oce. Further, they can provide contact information and establish
transactional relationships with the youth who visit their sites. Cam-
paigns can also shape candidate identity, establish candidate credibility,
and build community among the targeted youth.
Further, candidates can be strategic by taking advantage of this new
mode of direct campaigning on social media by addressing a more
complete range of policy positions than can be conveyed on television
or radio. It is unlikely that content on social media will be any more
misleading or deceptive than the standard campaign presentations are.
13. Ibid., 344.
14. Ibid., 345.
e presence of vigilant bloggers, Kenyans On Twitter (KOT) and social
media enthusiasts who are motivated to scrutinize every word posted
by candidates, is a powerful incentive for candidates to stay close to
facts. Unmediated communication better realizes voter independence,
the breadth and depth of policy debate, and candidates’ control over
their messages. ese gains may ultimately be signicantly augmented
by the collective benet of having more enthusiastic and engaged vot-
ers.
us, the challenge is not simply to identify new communication
practices and their eects on the content of election campaigns, but to
understand how encounters between technologies of communication
and political processes create new conditions for the formation of is-
sues of common interests and their publics.
Democratic Potentials of Social Media
Reections on the potential of online media for democracy began
immediately aer the internet was introduced. Advocates of the idea of
electronic democracy share a belief in the ability of new media to create
the conditions for political participation.15 While optimists see social
media as essentially progressive and improving the quality of election
campaigns, sceptics regard the claims made for social media as too
enthusiastic.16 erefore, the discussion of the democratic potential
of social media in this section shall be narrowed down to aspects of
campaigns in Kenyan elections, in so far as social media open spaces
for candidates’ engagement with voters in a way that traditional forms
of campaigns such as rallies, advertising, and mainstream media may
not.
15. Street, Mass Media, Politics and Democracy, 270.; C. D. Staton, “Democracy’s
Quantum Leap,” Demos Quarterly 3 (1994): 31–32.
16. Barbara Pfetsch and Silke Adam, “Democratic Potentials of Online Communi-
cation for Political Debate,” in Political Communication in the Era of New Technologies,
ed. Boguslawa Dobek-Ostrowska and Jan Garlicki (Frankfurt: Peter Lang, 2013), 31.
110
SOCIAL MEDIA & POLITICAL CAMPAIGNS IN KENYA | John O. Ndavula
111
Chapter 5 The Political Impact of Social Media
e democratic potential of social media is reected in its open ac-
cess for candidates and the potential for interactivity and coexistence
of horizontal and vertical communication with voters. If we consider
the structure of mainstream media, we note that the limited carrier
capacity and the logic of news factors—such as social status, political
oce, or prominence—makes the latter have an inherent selection bias
that privileges elite candidates access to media.17 Mass media’s selec-
tion bias produces a cumulative inequality. is means that candidates
who depend the most on mass media to gain political inuence have
the most dicult barriers to its entry and access.18 On the other hand,
social media would change the set of actors and voices in public debate
on election, with obvious consequences for political decision-making
and mobilization. From a normative point of view, the inclusiveness of
public debate is a fundamental requirement of democracy.19
It is against this background that I examine the potential of social
media to make up for this weakness of mainstream media and whether
social media oer new opportunities for political campaigns. us,
the question is whether social media are challenging traditional
mass-media–elite linkages by oering candidates without easy access
to public debate new venues for recognition. Social media could be
powerful enough to compensate for cumulative inequality, and occa-
sion changes in contemporary political communication.20 Social media
could provide a forum for a truly free exchange of ideas and views, un-
constrained by imbalances of power and resources. Technology change
brings with it new ideas and possibilities, and new notions of democ-
racy. e politics of e-democracy are also the politics of technology,
and both are tied intimately to the fact that forms of communications
are also systems of power. To the extent that social media campaigns
17. Ibid., 32.
18. Gadi Wolfsfeld, Media and Political Conict. News from the Middle East (Cam-
bridge: Cambridge University Press, 1997), 24.
19. Pfetsch and Adam, “Democratic Potentials of Online Communication,” 33
20. Ibid., 32.
contribute to the shi of power from one set of individuals to another,
we can say that social media have changed the power structure and
organizational culture within campaigns.21
Even then, the adoption of social media for political campaigns
attracts some criticism. Sceptics fear the impact of social media on
political campaigns, seeing it as a turn for the worse. ey argue that
although social media enhance the ow of information, which is
a central tenet of democracy, it does not follow that all information
enhances democracy. For example, voters may seek information on
social media that arms their already shared political beliefs rather
than the information that challenges them. Moreover, full participa-
tion via social media is generally hampered by problems such as time,
size, knowledge, and access.
In addition, there is a danger that social media may reproduce the
same kind of inequalities witnessed in mainstream media and hence
undermine democracy. Established political actors may prot from
the makeup of the internet because they have resources to exploit
online communication.22 Entrenched political interests will merely use
the new online aordabilities to their advantage, acting to sustain the
status quo rather than change it. In this sense then, technology will end
up serving the interest of those with power. However, as everyone can,
in principle, use the web to address the public, the hope remains that
previously marginalized actors are able to leverage the power of social
media to gain visibility among voters.23
While these observations need not be ignored, others hold the view
that technology is autonomous, and that it is directed by a technical
rationality that is not only independent of centres of political power,
but actually dictates to them. at is, we can do little more than accept
the inevitable technology and subsequent social revolution. is ac-
count of technology rests on a strong notion of determinism. But it is
21. Foot and Schneider, Web Campaigning, 195.3
22. Pfetsch and Adam, “Democratic Potentials of Online Communication,” 33.
23. Ibid., 33.
112
SOCIAL MEDIA & POLITICAL CAMPAIGNS IN KENYA | John O. Ndavula
113
Chapter 5 The Political Impact of Social Media
important to remember that the same technology can have a dierent
meaning and dierent eects according to its place in space and time
and the culture within which it operates.24
Candidates campaigning on social media have the power to dene
and extend a campaign organization, as well as to lter, destroy, or pro-
tect information for the campaign organization.25 e way information
is presented and organized on social media is correlated with forms
of political discourse. In other words, the citizen’s capacity to make
political judgements is dependent upon the way in which political
information is delivered and received.26
It is worth noting that when social media are linked to other political
actors they can facilitate quasi-independent interactions among po-
litical supporters. In addition, campaigns utilizing social media cede
control over the messages to which visitors may be exposed when they
follow links to sites produced by other actors. Meanwhile, voters can
create and recreate certain messages, multiplying the possible contexts
and interpretations of the original message. Ultimately, campaigns
may face increased demands of accountability from the public because
the web content is more permanent.
Instead of being forced to accept one of these competing views
of technology, it would be fair to say that technology structures our
choices and preferences, but not in a wholly deterministic way. We can
see voters’ relationship to social media and the meaning attached to
its contents as being created in the material and social context of their
lives. e way social media are used, and the signicance of the mes-
sages it delivers, are contingent.27 e implication then is that we need
to see how systems of communication construct dierent opportuni-
ties for political engagement. ere is some evidence at this point that
social media campaigning restructures the practice of politics, and in
24. Street, Mass Media, Politics and Democracy, 277.
25. Howard, New Media Campaigns and the Managed Citizen, 168–169.
26. Street, Mass Media, Politics and Democracy, 279.
27. Ibid., 278.
so doing, reshapes our democratic practice. In the end, though, it may
matter less what tools are used but rather how eectively and creatively
the tools are used.
Conclusion
It is useful to examine the contributions of social media to voting
outcomes, with an eye towards understanding that adoption rates are
inuenced by the success of an innovation. Our discussion has dem-
onstrated that political candidates perceive social media as having a
moderate potential to signicantly contribute to election outcomes in
Kenya, while at the same time they contend that social media have
made signicant contributions to voting decisions in Western democ-
racies. Candidates, therefore, opined that there were factors that con-
tributed to electoral success other than the medium used for political
campaigns. ese factors include political party, candidates’ personal
character, policy preferences, ethnicity, gender, education, and oc-
cupation, among others. Take the variable of ethnicity, for instance:
Kenya, being a polarised country, grapples with issues such as negative
ethnicity, which may impinge on voting decisions. In this case then,
although social media increase the content that is available for voters,
the voters themselves may turn to biased but favoured sites, hence wall-
ing themselves o from alternative opinions. Although such setbacks
exist, social media campaigns necessarily restructure the practice of
politics and in so doing reshape our democratic practice. Social media
are likely to have the greatest impact on younger Kenyans, who tend
to be more adept with technology, and to have integrated them more
closely into their personal lives. As social media spread even further
throughout society, we may expect them to play a pivotal role in elec-
toral campaigns in Kenya.
115114
SOCIAL MEDIA & POLITICAL CAMPAIGNS IN KENYA | John O. Ndavula
Table 5.1: Social Media Use in 2007 and Election Outcome
Electoral Ofce (%) Chi-square
test
President
(n=4)
Governor
(n=25)
Senator
(n=31)
Women Rep.
(n = 34)
MP
(n=211)
Strongly
disagree 0.0 8.0 9.7 11.8 6.2
X2=7.785
df = 16
Sig = 0.955*
Disagree 0.0 28.0 29.0 29.4 29.4
Neither agree
nor disagree 50.0 44.0 29.0 26.5 28.9
Agree 50.0 20.0 22.6 23.5 30.3
Strongly agree 0.0 0.0 9.7 8.8 5.2
* Correlation is signicant at the 0.05 level.
Table 5.2: Use of Social Media and Election Outcome
Electoral Ofce (%)
Chi-square
test
President
(n=4)
Governor
(n=25)
Senator
(n=31)
Women Rep.
(n=34)
MP
(n=211)
Strongly
disagree
0.0 4.0 6.5 8.8 0.9
X2 = 19.809
df = 16
Sig=.229*
Disagree 0.0 12.0 6.5 11.8 9.0
Neither agree
nor disagree
0.0 24.0 38.7 8.8 13.3
Agree 75.0 52.0 41.9 58.8 64.5
Strongly agree 25.0 8.0 6.5 11.8 12.3
* Correlation is signicant at the 0.05 level.
CHAPTER 6
Summing Up Social Media and Election Campaigns
e primary aim of writing this book has been to provide an account
of the initial period of the adoption of the social media campaign in
Kenyan electoral politics. e analysis herein has pointed towards a
wide range of signicant impacts of social media on the electoral pro-
cesses in Kenya. e book has explored how and why this shi has oc-
curred, with an emphasis on how candidates executed their individual
social media campaigns.
I began from the premise that, while much is known about the
existence of social media as a platform for political campaigns, little is
known about how social media are reconstituting and reshaping politi-
cal communication in Kenya. In chapters 1 and 2, I charted the history
of party political communication in Kenya from the pre-independence
period to the present. I mapped the key changes characteristic of the
pre-modern, modern and post-modern campaign periods. I further
focused on the intersection between social media and election cam-
paigns. is was followed by a discussion on the political implications
of these new modes of communication.
116
SOCIAL MEDIA & POLITICAL CAMPAIGNS IN KENYA | John O. Ndavula
117
Chapter 6 Summing Up Social Media and Election Campaigns
In chapters 3 and 4, I gave a theoretical perspective on the adoption
of social media by reviewing diusion of innovations theory. Further,
I considered the adoption of social media in the 2013 elections at the
presidential, gubernatorial, senatorial, women representative, and
parliamentary levels. e factors that I considered critical for adoption
were candidate characteristics, technological factors, and the environ-
ment. I probed deeper into candidates’ online behaviour by examining
their techniques of social media production for election campaigns
in Kenya. I assessed how the four production techniques (informing,
involving, connecting, and mobilizing) were and were not utilized in
the 2013 election campaigns in Kenya.
In chapter 5, I examined the political impact of social media and
assessed what the stakes are for Kenya. I explored the eectiveness of
social media as channels for electoral campaigns in Kenya and consid-
ered the prospects and challenges of using social media to reach the
younger age demographic.
To sum up, the central argument of the book is that social media
have come to play an increasingly important role in politics in Kenya
in the post-colonial era. e analysis presented herein connects the
many issues and arguments that are raised by the relationship between
social media and politics. e book not only oers new insight into
the current patterns of social media use, but also sheds light on the
emerging patterns of political communication, which in turn inuence
the practice of political campaigns in Kenya.
Looking Forward to the Future
Based on the developments witnessed in the 2007 and 2013 elec-
tions, and on the current trends in technology innovation and use,
signicant changes can be anticipated in campaigns on social media
over the next few elections in Kenya.
Regarding adoption of social media, it is likely that candidates will
employ social media for a wider variety of purposes. Given the oppor-
tunity for greater access to voters, candidates will seize the chance to
talk about their accomplishments and positions on issues. Candidates
will turn to Facebook pages and Tweets to highlight their most recent
accomplishments, to “score points” with voters. Formerly marginalized
candidates are likely to adopt a technology-driven system of direct or
unmediated political communication, to bypass the news media and
communicate directly with voters. is shi holds open the possibility
of improving voters’ ability to make decisions based on the candidates’
issue positions, creating a more well-informed electorate.
A signicant dierence in election campaigning is likely to be the
reduced cost of communicating with voters. Indeed, reaching vot-
ers via social media is far less expensive than mounting a full-scale
advertising campaign. e lower entry costs could potentially make
for a larger pool of candidates, a more multisided ow of information,
and eventually, more competitive elections. As Kenya enters the post-
modern period, its electoral process could well be rejuvenated.
Social media makes possible more personalized forms of candidate-
to-voter communication. Voters can gain access to information that
is more substantive and personally relevant than that provided by
mainstream news organizations’ coverage of the campaign. Social
media aord voters an opportunity to escape the sideshow of combat
between the candidates and the press.
Going forward, the act of establishing social media sites will become
more regular, raising the baseline of political information available to
citizens online. However, campaign teams may need to continuously
identify the most popular social media platforms with Kenyan voters
and invest in political marketing through those platforms. A case in
point is the recent WhatsApp messaging application, which is gaining
momentum in Kenya. is platform, and other technologies of its kind,
will enable campaigns to engage in greater segmentation of the audi-
ence and more rened targeting of their messages and informational
resources.
118
SOCIAL MEDIA & POLITICAL CAMPAIGNS IN KENYA | John O. Ndavula
119
Chapter 6 Summing Up Social Media and Election Campaigns
In the coming election cycles, we can expect that the techniques of
using social media for election campaigns will change modestly. Some
online practices such as informing will remain. Other techniques like
involving, connecting, and mobilizing will likely continue to evolve
gradually over time. Campaigns will likely seek to coproduce more
content in response to the challenge of producing greater quantities
of fresh and interesting information. However, we expect that any new
social media technologies which emerge will also feature evolving
web production techniques. Unexplored social media functions like
resource mobilisation may be considered, subsequent to which legisla-
tion on political fundraising, particularly through accessible mobile
money transfer systems like M-Pesa and Airtel Money, will need to be
craed.
Online networks have been found useful in promoting permanent
campaigns.1 e structure of online technologies essentially allows
candidates to develop digital resources that can be migrated across
elections, allowing the permanent campaign to emerge. Campaigns
will begin informing voters earlier in the electoral season, and continue
long aer the election, contributing to the concept of the permanent
campaign.
Regarding the social media skills gap among candidates, those
candidates with less experience in this area would do well to seek out
specialised training on political marketing through social media. is
will enable candidates and their social media site managers to eec-
tively leverage social media as a campaign platform. e training will
help political actors to use relevant social media techniques to amplify
their political messages.
A major concern associated with political campaigning on social
media is the level of penetration of and access to the internet in Kenya.
Even though internet penetration has continued to grow steadily, for
most people it remains expensive and out of reach. e regulators,
specically the Communications Authority of Kenya and the relevant
1. Pfetsch and Adam, “Democratic Potentials of Online Communication,” 34.
government ministries and departments, ought to ensure broader ac-
cess to the internet. Moreover, it is important for political parties as
well as individual candidates to develop a social media campaign ac-
tion plan that will harness the power of social media for campaigns as
well as to supplement oine campaign eorts. For example, political
parties need to facilitate candidates’ acquisition of computers and other
handheld devices with high internet speeds for successful launching of
the online campaign.
Of course, campaigns that seek to capitalize on the benets of new
technology and increase their web presence must also be aware of the
inevitable risk of losing control. One of the most celebrated products
of the internet is the proliferation of online political discourse in the
form of user-driven content—a sign of a large, active, and engaged
segment of the electorate. But the degree to which campaigns can
control their message, while encouraging democratic participation on
the web, is a central concern of every social media election strategy.
Candidates should be concerned that their campaigns on social media
do not degenerate into avenues for propaganda for war, incitement to
violence, hate speech or advocacy of hatred. Indeed, candidates face
many new and obscure challenges in the digital realm and must walk
the line carefully when perfecting their internet strategies, while at the
same time coexisting with a community of empowered users who can
ably use the same tools against them.
Perhaps the biggest challenge that campaigns face in the modern
period is keeping up with the sheer pace of technological advancement
and change, which is unlikely to slow down any time soon. ere has
been a massive overhaul of the campaign landscape. is is only likely
to continue, and perhaps even to intensify. Campaigns must therefore
remain vigilant and strive to reach the equilibrium between embracing
interactivity and digital innovation, and maintaining control over their
campaign messages.
121
120
SOCIAL MEDIA & POLITICAL CAMPAIGNS IN KENYA | John O. Ndavula
Conclusion
e chapters in this volume suggest that new media oer many new
opportunities to political campaigns in Kenya, but there are limits to
what can be accomplished online. Even as candidates embrace technol-
ogy, traditional electioneering tactics should not be neglected. Social
media campaign tactics are best used in combination with traditional
methods like rallies, door-to-door canvassing, direct mail, and TV or
radio advertisements, since these conventional strategies are still very
useful in eliciting support for candidates in Kenyan elections. Cam-
paigns need, therefore, to embrace a hybrid system that incorporates
online and oine networking and campaign tools to stay aoat in
post-modern elections.
Online communication triggers new structures of political exchange
and may indeed fulll hopes for a more inclusive, democratic public
discourse on election campaigns. Indeed, technology is transform-
ing the shape and design of campaigns and elections, and campaigns
will need to adapt to take advantage of these new opportunities. e
current state of technology and social media foreshadows a future of
campaigns and elections in which the tech-savvy will be rewarded.
Candidates who embrace these advancements head-on will nd
themselves leading the pack, while those who do not risk being le
behind. Failure to acclimate to these technological shis may seriously
hinder any future campaign’s success. Candidates and campaigns must
be exible and responsive enough to adapt their strategies to the new
reality of a continuously evolving digital world.
BIBLIOGRAPHY
Abdelhay, Nawaf. “e Arab Uprising 2011: New Media in e Hands of a
New Generation in North Africa.” Aslib proceedings: New Information
Perspectives 64:5 (2012): 529–539.
Accounting Diary. “Insider Clue on Cellphone Success in Africa.” Ac-
cessed November 4, 2013. http://www.sowareaccountingadvice.
com/2010/04/insider-clue-on-cellphone-success-in.html
Anderson, Chris. e Long Tail: Why the Future of Business is Selling Less of
More. New York: Hyperion, 2006.
Balkin, Jack M. “Digital Speech and Democratic Culture: A eory of Free-
dom of Expression for the Information Society.” New York University
Law Review 79:1 (2004): 1–55.
Baran, Stanley J. and Dennis K. Davis. Mass Communication eory: Founda-
tions, Ferment, and Future. Boston: Wadsworth, 2009.
Baumgartner, Jody and Jonathan Morris. “Myfacetube Politics: Social Net-
working Websites and Political Engagement of Young Adults.” Social
Science Computer Review 28:24 (2010): 24–44.
Benoit, William L. Communication in Political Campaigns. New York: Peter
Lang, 2007.
122 123
SOCIAL MEDIA & POLITICAL CAMPAIGNS IN KENYA | John O. Ndavula Bibliography
Bohler-Muller, Narnia and Charl Merwe. “e Potential of Social Media to
Inuence Socio-Political Change on the African Continent.” Africa
Institute of South Africa, brieng: 46, 2011.
Borja, Jordi and Manuel Castells. Local and Global. London: Earthscan, 1997.
Bosch, Tanja. “Youth, Facebook and Politics in South Africa.” Journal of
African Media Studies 5:2 (2013): 119–130.
Bourgault, Louise. Mass Media in Sub-Saharan Africa. Bloomington: Indiana
University Press, 1995.
boyd, danah m. and Nicole B. Ellison. “Social Network Sites: Denition, His-
tory, and Scholarship.” Journal of Computer-Mediated Communication
13:1 (2007): 210–230.
Castells, Manuel. “Communication, Power and Counter-Power in the Net-
work Society.” International Journal of Communication 1:1 (2007):
238–266.
Castells, Manuel. “Informationalism, Networks, and the Network Society: A
eoretical Blueprint.” In e Network Society: A Cross-Cultural Per-
spective, edited by Manuel Castells, 3–45. Northampton, MA: Edward
Elgar Publishing, 2004.
Castells, Manuel. e Rise of the Network Society. Oxford: Blackwell, 1996.
Chen, Peter. “Adoption and Use of Digital Media in Election Campaigns:
Australia, Canada and New Zealand.” Public Communication Review,
1 (2010): 3–26.
Cogburn, Derrick and Fatima Espinoza-Vasquez. “From Networked Nomi-
nee to Networked Nation: Examining the Impact of Web 2.0 and Social
Media on Political Participation and Civic Engagement in the 2008
Obama Campaign.” Journal of Political Marketing 10 (2011): 189–213.
Communications Commission of Kenya. “Mobile Penetration in the Country
Continues to Increase.” Accessed November 6, 2016. http://www.cck.
go.ke/mobile/news/index.html?nws=/news/2013/Mobile_penetration.
html
Cook, Catherine. “Mobile Marketing and Political Activities.” International
Journal of Mobile Marketing 5:1 (2010): 154–163.
Cooper, Joel and Kimberlee D. Weaver. Gender and Computers: Understand-
ing the Digital Divide. Mahwah, NJ: Princeton University, 2003.
Croteau, David and William Hoynes. Media/Society: Industries, Images, and
Audiences. 5th ed. ousand Oaks, CA: Pine Forge Press, 2013.
Elmer, Greg, Ganaele Langlois and Fenwick McKelvey, e Permanent Cam-
paign: New Media, New Politics. New York: Peter Lang, 2012.
Evans-Cowley, Jennifer and Justin Hollander. “e New Generation of Public
Participation: Internet-Based Participation Tools.” Planning Practice &
Research 25:3 (2010): 397–408.
Farrell, David and Paul Webb. Political Parties as Campaign Organizations. In
Parties without Partisans: Political Change in Advanced Industrial De-
mocracies, edited by Russell Dalton and Martin Wattenberg, 103–125.
Oxford: Oxford University Press, 2000.
Fischer, Claude. America Calling: A Social History of the Telephone to 1940.
Berkeley: University of California Press, 1992.
Foot, Kirsten. A. and Steven M. Schneider. Web Campaigning. Cambridge:
MIT Press, 2006.
Foster, Steven. Political Communication. Edinburgh: Edinburgh University
Press, 2010.
Freedom House. “Freedom on the Net 2016: Kenya Country Prole.” Accessed
January 10, 2017. https://freedomhouse.org/report/freedom-net/2016/
kenya
Frère, Marie-Soleil. Elections and the Media in Post-Conict Africa: Votes and
Voices for Peace? London: Zed Books, 2011.
Greyes, Natch. “e Untapped Potential of Social Media: A Primer for Savvy
Campaigners.” Campaigns and Elections 300 (2011): 45–47.
Gulati, Girish. J. and Christine B. Williams. “Closing Gaps, Moving Hurdles:
Candidate Website Communication in the 2006 Campaigns for Con-
gress.” Social Science Computer Review 25:4 (2007): 443–465.
Gulati, Girish. J. and Christine B. Williams. “Social Media and Campaigns
2012: Developments and Trends for Facebook Adoption.” Social Sci-
ence Computer Review 31:5 (2013):577–588.
124 125
SOCIAL MEDIA & POLITICAL CAMPAIGNS IN KENYA | John O. Ndavula Bibliography
Gustafsson, Nils. “e Subtle Nature of Facebook Politics: Swedish Social
Network Site Users and Political Participation.” New Media & Society
14:7 (2012): 1111–1127.
Herrnson, Paul. S., Atiya K. Stokes-Brown and Mathew Hindman. “Campaign
Politics and the Digital Divide: Constituency Characteristics, Strategic
Considerations, and Candidate Internet Use in State Legislative Elec-
tions.” Political Research Quarterly 60:1 (2007): 31–42.
Homan, Auren. “e Social Media Gender Gap.” Accessed January 23, 2015.
http://www.bloomberg.com/bw/stories/2008-05-19/the-social-media-
gender-gapbusinessweek-business-news-stock-market-and-nancial-
advice
Hornsby, Charles. Kenya: A History Since Independence. London: I. B. Tauris,
2012.
Howard, Philip N. New Media Campaigns and the Managed Citizen. New
York: Cambridge University Press, 2005.
Howard, Philip N. and Malcolm R. Park. “Social Media and Political Change:
Capacity, Constraint and Consequence.” Journal of Communication 62
(2012): 326–359.
Humanipo. “Kenya ICT Board Launches Julisha ICT Survey Report 2013.”
Accessed September 3, 2016. http://icta.go.ke/pdf/Julisha%20Final%20
Report%201.pdf
Hyden, Goran and Michael Leslie. “Communications and Democratization
in Africa.” In Media and Democracy in Africa, edited by Goran Hyden,
Michael Leslie, and Folu F. Ogundimu, 1–28. New Jersey: Transaction
Publishers, 2002.
Internet World Stats. “Africa.” Accessed January 6, 2017. http://www.internet-
worldstats.com/africa.htm#ke
Internet World Stats. “e internet Usage Statistics for Africa.” Accessed
December 9, 2016. http://www.internetworldstats.com/stats1.htm
ITU. Measuring the Information Society. Geneva: ITU, 2013.
Iyengar, Shanto. Media Politics: A Citizens’ Guide. 3rd ed. New York: W. W.
Norton, 2016.
Jerey, Simon. “Ushahidi: Crowdmapping Collective that Exposed Kenyan
Election Killings.” Accessed October 18, 2016. https://www.theguard-
ian.com/news/blog/2011/apr/07/ushahidi-crowdmap-kenya-violence-
hague
Jesuit Hakimani Centre. Fencing Duty Boundaries: Whose Values Matter in
Kenya’s Devolved System? Nairobi: Jesuit Hakimani Centre, 2015
Jue, Arthur L., Jackie Alcalde Marr and Mary Ellen Kassotakis. Social Media
at Work: How Networking Tools Propel Organizational Performance.
San Francisco: Jossey-Bass, 2010.
Kamarck, Elaine C. “Political Campaigning on the Internet: Business as
Usual?” Pages 81–103 in Governance.Com: Democracy in the Informa-
tion Age. Edited by Elaine C. Kamarck, and Joseph S. Nye. Washington,
DC: Brookings Institution, 2002.
Kanyinga, Karuti. “Contestation over Political Space: e State and the
Demobilisation of Opposition Politics in Kenya.” Pages 39–90 in e
Politics of Opposition in Contemporary Africa. Edited by A. O. Oluko-
shi. Uppsala: Nordiska Afrikainstitutet, 1998.
Karimi, Faith. “Kenyan Chief Foils Robbery via Twitter, Highlights Reach
of Social Media.” Accessed October 12, 2016. http://www.cnn.
com/2012/02/18/world/africa/kenya-tweeting-chief/
Kavanagh, Dennis. “New Campaign Communications: Consequences for
Political Parties.” Harvard International Journal of Press and Politics 1:3
(1996): 60–76.
Kavanagh, Dennis. Election Campaigning: e New Marketing of Politics.
Oxford: Blackwell Publishers, 1995.
Kenya ICT Board. “Digital Kenya: A Study to Understand the On-line Life
of Kenyans.” Accessed December 3, 2015. https://www.slideshare.net/
tandaaKENYA/study-of-the-online-life-of-kenyans-2010
Kimani, Mwaura and Christine Mungai. “Campaign Finance: Price Tag of
Kenya 2012 Presidential Race to Hit $130 Million.” Accessed July 6,
2014. http://www.theeastafrican.co.ke/news/Campaign-nance-
-Price-tag-of-Kenya-2012-presidential-race/-/2558/1320582/-/
item/0/-/a9r31ez/-/index.html
126 127
SOCIAL MEDIA & POLITICAL CAMPAIGNS IN KENYA | John O. Ndavula Bibliography
Koster, Josh. “Long-Tail Nano Targeting: Al Franken’s Online Ad Buys Earned
an Unbelievable Return on Investment.” Politics Feb (2009): 23–26.
Kotler, Philip, Ned Roberto and Nancy Lee. Social Marketing: Improving the
Quality of Life. ousand Oaks, CA: Sage, 2002.
Kushin, Matthew and Masahiro Yamamoto. “Did Social Media Really Matter?
College Students’ Use of Online Media and Political Decision Making
in the 2008 Election.” Mass Communication and Society 13 (2010):
608–630.
Kwon, T. H., & Zmud, R. W. “Unifying the Fragmented Models of Informa-
tion Systems Implementation.” In Critical Issues in Information Systems
Research, edited by Richard J. Boland and Rudy Hirshheim, 227–251.
New York: John Wiley, 1987.
Laverty, Alex. “ICT, Social Media and Elections in Africa: A Prospective
Study.” Accessed July 22, 2017. https://theafricanle.com/ict/ict-social-
media-and-elections-in-africa-a-prospective-study/
Lees-Marshment, Jennifer. “e Marriage of Politics and Marketing.” Political
Studies 49 (2001): 692–713.
Maina, Wangui. “e money factor in race for Kenya’s top job.” Accessed
August 8, 2014. http://www.africareview.com/Special-Reports/
The-money-factor-in-race-for-Kenyas-top-job/-/979182/1708124/-
/9v0n5/-/index.html
Makama, Rimini. “Electronic Campaigns And Social Media Elections, e
Frontier Of Smart Politics.” Accessed June 22, 2017. https://medium.
com/organizer-sandbox/electronic-campaigns-and-social-media-
elections-the-frontier-of-smart-of-politics-47db04155405
Makinen, Maarit and Wangu M. Kuira. “Social Media and Postelection
Crisis in Kenya.” e International Journal of Press/Politics 13:3 (2008):
328–335.
Marriott, Michel. “Blacks Turn to Internet Highway, and Digital Divide Starts
to Close,” New York Times, March 31, 2006.
Masilo, Bontle and Batlang Seabo. “Facebook: Revolutionising Electoral Cam-
paign in Botswana?” Journal of African Elections 14:2 (2015):110–129.
Menya, Walter. “IEBC Caps Political Aspirants, Party Expenditure in 2017
Polls.” Accessed August 11, 2016. http://www.nation.co.ke/news/
IEBC-caps-political-aspirants-party-expenditure-in-2017-polls/1056-
3340778-sxtd31z/index.html
Monge, Peter. R. and Noshir S. Contractor. eories of Communication Net-
works. Oxford: Oxford University Press, 2003.
Mukhongo, Lynete L. and Juliet W. Macharia. Political Inuence of the Media
in Developing Countries. USA: IGA, 2016.
Mulgan, Geo. J. Communication and Control: Networks and the New Econo-
mies of Communication. New York: Guilford Press, 1991.
Murphy, Alix. “e Greatest Innovator in Mobile Payments Isn’t Apple.”
Accessed October 11, 2016. http://betanews.com/2015/04/27/the-
greatest-innovator-in-mobile-payments-isnt-apple/
Murthy, Dhiraj. Twitter: Social Communication in the Twitter Age. Cambridge:
Polity, 2013.
Ndavula, John O. and Hellen K. Mberia. “Social Networking Sites in Kenya:
Trigger for Non-institutionalized Democratic Participation.” Interna-
tional Journal of Business and Social Science 3:13 (2012): 300.
Ndavula, John O. and Joy Mueni. “New Media and Political Marketing in
Kenya: e Case of 2013 General Elections.” International Journal of
Arts and Commerce 3:6 (2014): 69–84.
Norris, Pippa. A Virtuous Circle: Political Communication in Post-Industrial
Societies. New York: Cambridge University Press, 2001a.
Norris, Pippa. Digital Divide? Civic Engagement, Information Poverty and the
internet Worldwide. New York: Cambridge University Press, 2001b.
O’Connor, Brendan, Ramnath Balasubramanyan, Bryan R. Routledge and
Noah A. Smith. “From Tweets to Polls: Linking Text Sentiment to Pub-
lic Opinion Time Series.” Accessed December 9, 2016. http://www.aaai.
org/ocs/index.php/ICWSM/ICWSM10/paper/viewFile/1536/1842
Odinga, Christa. “Use of New Media During the Kenya Elections.” Accessed
March 16, 2015. http://www.diva-portal.org/smash/get/diva2:633138/
FULLTEXT01.pdf
128 129
SOCIAL MEDIA & POLITICAL CAMPAIGNS IN KENYA | John O. Ndavula Bibliography
Orring, Frida. “Kenyan Elections 2013 – A Tech Fail or Not? Freedom on the
Internet.” Accessed December 3, 2015. http://blog.swedenabroad.se/
fxinternet/2013/05/10/kenyan-elections-2013-a-tech-fail-or-not/
Oyugi, Walter O., Peter Wanyande and C. Odhiambo-Mbai. e Politics of
Transition in Kenya: From KANU to NARC. Nairobi: Heinrich Boll
Foundation, 2003.
Parke, Phoebe. “How Many People Use Social Media in Africa?” Accessed
January 6, 2017. http://www.cnn.com/2016/01/13/africa/africa-social-
media-consumption/
Pfetsch, Barbara and Silke Adam. “Democratic Potentials of Online Commu-
nication for Political Debate.” Pages 31–42 in Political Communication
in the Era of New Technologies. Edited by Boguslawa Dobek-Ostrowska
and Jan Garlicki. Frankfurt: Peter Lang, 2013.
Pickard, Jim. “When Politics and Social Media Collide.” Accessed November
22, 2016. https://www..com/content/27a7d6c8-702f-11e6-a0c9-
1365ce54b926
Portland, “How Africa Tweets.” Accessed January 15, 2017. http://portland-
communications.com/publications/how-africa-tweets-2015/
Reuters. “Facebook Rakes in Users in Nigeria and Kenya, Eyes Rest of
Africa.” Accessed January 6, 2017. http://www.reuters.com/article/us-
facebook-africa-idUSKCN0RA17L20150910
Robinson, Les. “A summary of Diusion of Innovations.” Accessed Decem-
ber 9, 2016. http://www.enablingchange.com.au/Summary_Diu-
sion_eory.pdf
Rogers, Everett M. Diusion of Innovations. 5th ed. New York: Free Press, 2003.
Römmele, Andrea. “Political Parties, Party Communication and New Infor-
mation and Communication Technologies.” Party Politics 9:1 (2003):
7–20.
Scammell, Margaret. Designer Politics: How Elections are Won. Basingstoke:
Macmillan Press, 1995.
Schlozman, Kay L., Sidney Verba and Henry E. Brady. “Weapon of the Strong?
Participatory Inequality and the internet.” Perspectives on Politics, 8:2
(2010): 487–509.
Shaha, Abigail. “Long Tail Politics: Sell Less of More.” Politics Nov (2008):
14–15.
Slotnick, Allison. “‘Friend’ the President: Facebook and the 2008 Presidential
Election.” Pages 249–271 in Politicking Online: e Transformation of
Election Campaign Communications. Edited by Costas Pangopoulos.
New Jersey: Rutgers, 2009.
Spitulnik, Debra. “Alternative Small Media and Communicative Spaces.”
In Media and Democracy in Africa. Edited by Goran Hyden, Michael
Leslie and Folu F. Ogundimu, 177–206. New Jersey: Transaction Pub-
lishers, 2002.
Staton, C. D. “Democracy’s Quantum Leap.” Demos Quarterly 3 (1994):
31–32.
Stiung, Friedrich E. Institutionalizing Political Parties in Kenya. Nairobi:
Friedrich Ebert Stiung, 2010.
Straw, Will. “Yes we did? What Labour learned from Obama.” In e Internet
and the 2010 Election: Putting the Small ‘p’ Back in Politics? Edited by
Rachel Gibson, Andy Williamson and Stephen Ward, 43–50. London:
Hansard Society, 2010.
Sudulich, Maria. L. and Mathew Wall. “‘Every Little Helps’: Cyber-campaign-
ing in the 2007 Irish General Election.” Journal of Information Technol-
ogy & Politics 7 (2010): 340–355.
Sudulich, Maria. L., Mathew Wall, Elmer Jansen and Kevin Cunningham. “Me
Too for Web 2.0? Patterns of Online Campaigning among Candidates
in the 2010 UK General Elections.” Accessed April 26, 2014. http://ipp.
oii.ox.ac.uk/sites/ipp/les/documents/IPP2010_Sudulich_Wall_Jans-
en_Cunningham_Paper.pdf
Tedesco, C. John, J. L. Miller and J. A. Spiker. “Presidential Campaigning
on the Information Superhighway: An Exploration of Content and
Form.” Pages 51–63 in e Electronic Perspective on the 1996 Campaign
Communication. Edited by Lynda Lee Kaid and Dianne G. Bystrom.
Mahwah, NJ: Lawrence Erlbaum, 1999.
130 131
SOCIAL MEDIA & POLITICAL CAMPAIGNS IN KENYA | John O. Ndavula Bibliography
Williams, Christine. B., and Girish J. Gulati. “Social Networks in Political
Campaigns: Facebook and the Congressional Elections of 2006 and
2008.” New Media and Society 15:1 (2012): 52–71.
Wolfsfeld, Gadi. Media and Political Conict. News from the Middle East.
Cambridge: Cambridge University Press, 1997.
Xenos, Michael A. and Lance W. Bennett. “e Disconnection in Online Poli-
tics: e Youth Political Web Sphere and US Election Sites, 2002–2004.”
Information, Communication, & Society 10:4 (2007): 443–464.
Tempere, Kaja. “A Walk in the Public Relations Field: eoretical Discus-
sions from a Social Media and Network Society Perspective.” Central
European Journal of Communication 1 (2011): 49–61.
Tornatzky, Louis. G. and Katherine J. Klein. “Innovation Characteristics and
Innovation Adoption Implementation: A Meta-Analysis of Findings.”
IEEE Transactions on Engineering Management 29:1 (1982): 28–45.
Tracey, Lauren. “ Will Social Media Inuence Election Campaigning in South
Africa?” Accessed June 22, 2017. https://issafrica.org/iss-today/will-
social-media-inuence-election-campaigning-in-south-africa
Trent, Judith. S., Robert V. Friedenberg and Robert E. Denton. Political
Campaign Communication: Principles and Practices. 7th ed. New York:
Rowman & Littleeld, 2011.
Utz, Sonja. “e (Potential) Benets of Campaigning Via Social Network
Sites.” Journal of Computer Mediated Communication 14 (2009):
221–243.
van Dijck, Jose. e Culture of Connectivity: A Critical History of Social Media.
New York: Oxford, 2013.
van Dijk, Jan. e Deepening Divide. ousand Oaks, CA: Sage, 2005.
Volkmer, Ingrid. “e Global Network Society and the Global Public Sphere.”
Development 46:1 (2006): 9–16.
Ward, Stephen and Rachel Gibson. “European Political Organisations and
the internet: Mobilisation, Participation, and Change.” In Routledge
Handbook of internet Politics. Edited by Andrew Chadwick and Philip
N. Howard, 25–39. New York: Routledge, 2009.
Wasserman, Ira M. and Marie Richmond-Abbott. “Gender and the internet:
Causes of Variation in Access, Level, and Scope of Use.” Social Science
Quarterly 86: 1 (2005): 252–270.
Widner, Jennifer A. e Rise of a Party-State in Kenya: From “Harambee!” to
“Nyayo!”. Berkeley: University of California, 1992.
Williams, Christine B. and Girish J. Gulati. “e Political Impact of Facebook:
Evidence from the 2006 Midterm Elections and 2008 Nomination
Contest.” Politics and Technology Review 1:1 (2008): 11–24.
133
INDEX
A
adopter characteristics xvii, 42,
44, 46, 68
Airtel Money 118
autocratic rule 9
B
Balkin, Jack 121
bandwidth xii
ban in campaigning 3
billboards 5
bloggers 109
Bourgault, Louise 122
C
campaign design 96
campaign nancing 13
campaign funds 93
candidate credibility 95, 108
candidate identity 95, 108
Castells, Manuel 122
censorship 29, 36
centralised campaign manage-
ment 7
citizen engagement xiv
commercials 5
communication professionals 7
compatibility xvi, 42, 44, 45, 53,
70
competitiveness of the race xvii,
42, 46, 60, 62, 70
complexity xvi, 42, 44, 45, 53,
55, 70
congressional races 106
constituency factors 34, 38, 67
constituency type xvii, 42, 60,
70
134 135
SOCIAL MEDIA & POLITICAL CAMPAIGNS IN KENYA | John O. Ndavula Index
K
KANU vii, 3, 8, 9, 10, 128
knowledge activism x, xi
knowledge gap xi
L
level of oce xvii, 42, 46, 48, 60,
61, 70
liberalization 6
literacy levels 64
localism 3
logistic regression model 68, 69
long campaign 10
long-tail marketing 39, 97
M
Makinen, Maarit 126
marginalized candidates 117
mass media xv, 1, 5, 11, 26, 29,
70, 98, 110
mature democracies 16
media management 10
media politics 1
media reception 19
micro-blogging 35, 56
mobile money transfer 93, 118
modern campaign 7, 9, 12, 15,
17, 38, 115
modern media xviii, 13
modern politics xviii, 5
M-Pesa 93, 118
multi-party 4, 6, 10, 12, 13
N
negative campaign 92, 106
News Feed 21
non-mediated communication
7
normative 110
O
Observability ii, xvii, 42, 53, 57
Odinga, Christa 127
oine 28, 59, 87, 88, 89, 93, 94,
98, 119, 120
online fundraising 94
online visitors 88, 98
Open Graph 21
opinion formation 23, 30
P
pamphlets 8
participatory rights 29
partisan press 7, 9
party aliation xvii, 42, 46, 60,
69, 70, 71
party defections 16
party organisation 23, 30
constitution 2, 3, 4, 10, 12, 36
content sharing xiii
D
debate 3, 5, 109, 110
decentralized campaign 16
democracy xiv, xvii, xviii, 8, 29,
30, 40, 67, 70, 98, 101, 109,
110, 111
democratic culture 28
democratic space 28
democratic system 2
demographics xvii, 22, 29, 42,
46, 60, 64, 69, 70
determinism 112
devolved system 2, 4, 12
dialogical communication 6
diusion of innovations xvi, 40,
41, 42, 43, 44, 63, 70, 116
diusion of social media 2, 36,
41
digital innovation 120
digital media 43, 87
digital natives 15, 16
digital technology 17
distributed networks 19
E
early adopters 45, 46, 52
e-democracy 110
email xv, 55, 88
environmental factors xvii, 42,
68
ethnic-based campaigning 3
ethnicity 3, 101, 113
ethnic voting blocs 104
F
bre-optic cable xii
yers 87, 88, 89
Foot, Kirsten 123
fragmentation of channels 38
G
Gender 48, 50, 52, 73, 75, 79,
80, 81, 83, 123, 124, 130
grassroots participation 10
Gulati, Girish 123
H
handout culture 8
harambee 8
hashtag 25
Hyden, Goran 124
I
Income Level 66, 78
Innovators 45, 46
136 137
SOCIAL MEDIA & POLITICAL CAMPAIGNS IN KENYA | John O. Ndavula Index
technology adoption model
xvii, 42
television 1, 5, 10, 13, 14, 87,
108
text messages 15, 90, 91
transactional relationships 95,
108
trialability 44, 45
tweets 25, 26, 27, 28, 128
U
user-generated content 20
V
vertical communication 110
Vision 2030 xii
W
web campaigning 43
websites 15, 24, 34, 97
personal gratication 107
political advertising 10, 12, 15
political consultants 15
political marketing 23, 38, 86,
96, 98, 117, 118
political mobilization 26, 27, 33
political power 10, 111
political transition xiii
posters 5, 35, 57, 88
post-independence 3
post-modern campaign 15, 38,
115
pre-modern campaign 7
presidential campaign xii, 65,
91, 106
propaganda 12, 57, 91, 92, 93,
119
public gatherings 5
public opinion 58
public sphere 28
Q
qualitative approach xv
quantitative approach xiv
R
radio 1, 5, 11, 13, 57, 87, 108,
120
rallies 5, 8, 15, 16, 30, 32, 39, 40,
59, 91, 93, 110, 120
relative advantage xvi, 42, 44,
45, 53, 70
retail politics 7
risk-averse organizations 96
Rogers, Everett 128
S
selection bias 110
single-party 3, 6, 9, 12
socialised communication 1
social marketing 65, 92, 96, 99
social media production xvii,
116
social media site managers xv,
56, 70, 86, 91, 104, 118
social revolution 112
social system xvii, 42, 43, 46, 60
social ties 19
statutory controls 12
T
technical expertise 35, 55, 56,
85, 95
technical rationality 111
techniques of informing 85
technological advancement 119
technological development xii
technological factors xvii, 42,
69, 116
