top of page

The Impact of Electoral Violence on Electoral Participation and Support for Democracy in Kenya

Martin Fikiri Oswald

1. Introduction

Electoral violence is a common occurrence in Africa. Most African countries have experienced varying degrees of electoral violence. Citizens’ fear and experiences of electoral violence have had an impact on their support for democracy on the one hand and their propensity to participate in democratic activities such as voting and attending campaign meetings on the other. Thus, it is the purpose of this study to determine the extent to which fear of electoral violence impacts individual support for democracy and participation in electoral activities in Kenya. I


t is hypothesized that individuals who have experienced and fear some form of electoral violence and/or intimidation are less likely to support democracy as well as less likely to participate in democratic activities such as voting and attending campaign meetings. Election campaigns are the most essential platforms for politicians to connect and communicate with voters about their election manifestos and goals. Thus, nonparticipation in election campaigns, especially in newly democratizing countries, is not a good indicator for democratic consolidation. Correspondingly, nonparticipation out of fear of violence is likely to lead to a communication breakdown between candidates and the electorate, which is harmful, especially to participatory democracy. The viability of democracy largely depends on widespread citizen participation without which the quality of elected leaders is likely to be negatively affected.


2. Electoral democracy in Kenya

In the early 1990s, Africa rushed to reintroduce “liberal democracy”, defined in terms of multiparty politics, the conduct of regular and competitive elections among parties and candidates, separation of powers between and among the branches of government, checks and balances, an independent judiciary and rule of law, an independent and impartial election management bodies, civil society organizations as well as an impartial and balanced media. These are seen as the necessary ingredients of a successful democracy. Huntington (1993) referred to this rush as “the third wave of democracy,” which surged over Africa after decades of various forms of autocracy such as one-party regimes, military rule, and dictatorships. Kenya was not left behind in this stampede. After more than 30 years of monopolistic single-party administration, Kenya transitioned to democracy in 1991.


Prior to the transition, Kenya African National Union (KANU) solely governed for more than three decades. After the opening of the political space, Kenya saw a proliferation of political parties and related democratic institutions. Following this transition, in 1992 the country held its first multiparty elections with eight newly formed parties. Since then, Kenya has successfully and uninterruptedly held elections every five years. However, these elections have been marred by sporadic acts of violence of varying degrees, with the 2007 elections being the most violent.


Apart from conducting elections uninterruptedly for three decades and providing the citizens with the democratic right of electing their leaders and the right to be elected for political offices, the restoration of democracy in Kenya has been successful in several other ways. Through the conduct of elections, there have been successful transfers of power between and among political leaders and political parties. Along with that, there has been an improvement in the quality of elections in terms of the rules and laws governing the practice. There has also been an improvement in the number of people who participate in the activity such as the voters (voter turnout) , candidates vying for political offices, political parties, civil society organizations and institutions. All these have increased in quality and quantity over time making elections increasingly competitive.


However, the transition to democracy has not been without difficulties and challenges. Election-time clashes and disputes between rival candidates and party supporters have often sparked electoral acts of violence. Out of apprehension of violence, voters tend to avoid attending campaign events and are more likely to skip the polls on election day. This has resulted not only in low voter turnout but also in minimal support for democracy. This has been the trend not only in Kenya but in other African countries, as has been documented and evidenced by Mpabanga (2000) and by Kuenzi and Lambright (2007). And, as Adejumobi (2000:59) puts it, elections have been the main victim of the common occurrences of electoral violence across Africa.


According to studies on electoral violence in Africa, more than 50 percent of elections since the transition in the 1990s can be characterized as violent, with the citizens suffering from abuse, intimidation, property loss, and, in some cases, fatalities as a result of the electoral process (Burchard, 2015:3). As evidence of the seriousness of the issue, Lindberg (2006) states that between 1990 and 2000, there was electoral violence in 80 percent of the elections held in Africa. This is similar to recent investigations which show that almost no African election is free of violence (Straus and Taylor; 2009; Adolfo et al. 2012).


Given the severity of the problem, electoral violence has attracted massive attention among political scientists and sociologists with most scholars focusing on the causes and effects of electoral violence. Studies have identified the closeness of competition and institutional frameworks in the management of elections as some of the triggers of violence (Wilkinson 2006; Chaturvedi 2005; Kristine Höglund 2009; Straus and Taylor 2009; Collier 2011). On the effects of electoral violence, a sizeable literature focuses on voter turnout and democracy (Bratton 2008; Collier and Vicente 2014; Condra et al. 2018; Ley 2018; Burchard 2015, 2020). On the other hand, much less scholarship has attempted to examine the impact of electoral violence on other political attitudes. Those studies which have been done have concluded that electoral violence undermines citizens’ support for democracy and political trust (Burchard 2020; Höglund 2009; Opitz, Fjelde, and Höglund 2013).


3. Research design

The paper uses the violence in Kenya’s 2007 elections and the subsequent election in 2012 as a case study. The analysis was performed using Afrobarometer survey data conducted in 2014, i.e the survey after the 2012 elections. The choice of this data is to allow the study to assess how supportive or unsupportive of democracy citizens were after experiencing electoral violence. This is consistent with the study’s premise that citizens who are exposed to and/or have experienced electoral violence are less inclined to support democracy and hence may be unwilling to participate in democratic activities like voting and attending election campaign meetings. Consequently, citizens are likely to prefer less violent non-democratic regimes to democratic regimes that instill violence.


Self-reported individual fear of electoral violence is this study’s dependent variable. This is selected from Afrobarometer’s survey question where respondents were asked: “During election campaigns in this country, how much do you personally fear becoming a victim of political intimidation or violence?” Responses were coded (0) fear and (1) do not fear.


The independent variables are listed in Table 1 below.

4. Data analysis

Descriptive statistics using SPSS (Figure 1, Table 2 and Table 3) show that more than two third (67 percent) of the respondents indicated fear of electoral violence in the 2012 elections. This fear of electoral violence explains why almost three quarter (74 percent) of the respondents indicated that they did not attend election campaigns. After performing a frequency analysis, it was found that half of the respondents (50 percent) are not interested in politics and only less than a quarter (22 percent) indicated interest in politics. With regards to trusting the election management body - the Independent Electoral and Boundaries Commission (IEBC), only slightly lower than a quarter (23 percent) indicated trust for the body. The rest either do not trust it at all (28 percent) or have little trust (49 percent).


A simple crosstabulation of fear of electoral violence and selected variables provided useful results that substantiate the frequency analysis results. The crosstabulation results show that more than three quarter of respondents who fear electoral violence (76 percent) did not attend election campaign meetings. Although the voter turnout from the sample seems to have been relatively high at 80 percent, more of those who fear electoral violence did not vote (21 percent) compared to 17 percent of those who do not fear electoral violence. Similarly, there were fewer respondents that fear electoral violence who voted (79 percent) compared to 83 percent of the respondents who voted and indicated they did not fear electoral violence. Therefore, most respondents who fear electoral violence did not vote.


With regards to support for democracy, the simple crosstabulation of fear of electoral violence and support for democracy showed that those who fear electoral violence were less supportive of democracy (78 percent) than those who do not fear electoral violence (81 percent). The same applies to the trust variable where fewer respondents who fear electoral violence (19 percent) trust the electoral management body than 31 percent of the respondents who do not fear electoral violence. This is to say that those who fear electoral violence are less trustful of IEBC than those who are not fearful.

5. Implications and Conclusion

The results of this study have shown that electoral violence tends to discourage the electorate from taking part in electoral activities including voting and attending election campaign meetings. Similarly, the fear of electoral violence has proven to affect the electorate on other attitudinal and behavioural aspects of democracy apart from voting and attending election campaigns. These other attitudinal aspects include support for democracy, political trust and political interest. It was found that citizens who fear electoral violence are less likely to have an interest in politics, support democracy and trust institutions that manage elections.


These findings are congruent to those found by other scholars on the subject matter in Africa. They include Kramon (2013), Bratton, Bhavnani, and Chen (2012) Bratton, (2008), Burchard, (2015, 2020), Collier & Vicente (2014), Condra et al., (2018) and Ley, (2018) who concluded that electoral violence had a significant impact on the levels of electoral participation in Africa at varying levels and intensity. From the findings of this study, it is important to strengthen democratic institutions to earn the citizen’s trust and confidence. By so doing citizens’ support of democracy will be restored hence more participation which is important for the survival of democracy. With strong and trustworthy institutions in place – such as competent and impartial electoral management bodies and a vigorous independent media- we might expect to have more reliable and peaceful elections.


Aside from institutional shortcomings in administering peaceful elections, electoral violence is attributed to a variety of issues. The closeness of elections is cited as one of the factors contributing to violent elections. In Africa, proximity is usually related to and based on ethnicity, religiosity, and/ or regionalism, where contending parties want the election to be in their favour, triggering fights with opponents. For example, ethnicity is cited to be significantly responsible for the 2007 post-election violence in Kenya (Wahman 2014; Burchard 2015; Gutiérrez-Romero and LeBas 2015). Therefore, in addition to building democratic institutions, it is important to inculcate political tolerance in society and appreciation of the diversity that exists in the country.

Screen Shot 2023-11-16 at 6.44.22 PM.png
Screen Shot 2023-11-16 at 6.46.21 PM.png
Screen Shot 2023-11-16 at 6.48.23 PM.png
bottom of page