Voter Behaviour amongst Kenyan Youths in relation to Party Primaries and General Elections: A Focus on Students at Masinde Muliro University of Science and Technology during the 2017 Election Cycle

This research study was conducted in 2017 at the Masinde Muliro University of Science and Technology situated in Kakamega County, Kenya. It was motivated by the desire to gain insights into the political attitudes, inclinations and perceptions of Kenyan youths towards voting in party primaries and general elections. The study site was purposively selected. The Fisher’s formula of determining sample size was used to arrive at 338 respondents who were picked through the simple random sampling procedure. Observation, structured interviews, unstructured interviews, and focus group discussions were deployed. The exploratory research design was utilized given that it was a fact-finding endeavour. Data was analyzed through descriptive statistical analysis with the aid of SPSS. It was revealed that 52% of respondents had taken part in the 2017 party primaries, 47% did not, while 1% failed to respond. A majority of respondents (88%) indicated they would participate in the 2017 general elections. Most respondents (91%) had registered to take part in the election exercise. On the importance of voting in elections, an overwhelming majority of 97% argued in favour of participation, 2% against, while 1% failed to give a clear response. The study concluded that despite electoral hurdles most youths in Kenya faced, they still recognized the importance of expressing their democratic preferences during party primaries and general elections.


INTRODUCTION
Resnick and Casale (2011) observed that the youth had long represented a key constituency for electoral mobilization in Africa, and that nowadays as the continent faces a fast-growing youth population increase that is disproportionately burdened by unemployment and underemployment, capturing the votes of this demographic is becoming more vital than ever before. However, despite their numerical significance and the historical relevance of generational identities within the continent, quite little is in the public domain in regards to the political participation of Africa's youth. It is for this reason that the study seeks to gain insights into the political attitudes, inclinations and perceptions of youths in relation to their involvement in party primaries and general elections in order to improve on existing policies.
A policy brief by the International Institute for Democracy and Electoral Assistance (2015) pointed out that young people with ages ranging from 15 to 35 constituted one-third of Africa's population, which justifies looking into their participation in democratic processes. A multiplicity of factors influences the voter behaviour of individuals. Makori (2015), in a study aimed at establishing youth perception of and participation in electoral democracy in Kenya, postulated that youth mentality stood out as one of the greatest barriers to effective political participation given that the attitudes they bore were birthed at the family level and continued to build up as they got exposed to the same negative mentality by different social actors.
It is an undisputed fact that party primaries lay the foundation for general elections. Caramani (2011) observed that political parties were a result of the need for participants to cooperate in order to nominate candidates, help them attract potential voters, and organize voter turnout in favour of their candidates. Kramon (2009), in a study on vote buying and turnout in Kenya's 2002 elections, concluded that Kenyans who had been approached by a vote buyer were about 14 percentage points more likely to vote than those who had not. Wiese (2011) pointed out that it is generally accepted that the youth are generally apathetic towards political activities and that such apathy is evidenced in low voter turnout during elections, and that when a significantly large proportion of the voting population shows apathetic attitudes towards the processes that enable participation in democracy, it questions the extent to which such democracy can be consolidated.
The evolutionary theory of political behaviour by Smith, Larimer, Littvay and Hibbing (2007) attempted to explain voter behaviour; these scholars observed that one of the enduring features of modern representative democracy was people's dissatisfaction with political leaders thereby influencing either their participation or nonparticipation in the electoral process.

Study Area, Target Population, Unit of Observation, and Unit of Analysis
This study was conducted at Masinde Muliro University of Science and Technology situated in Kakamega Town, Kakamega County, off the Kisumu-Webuye road. The site was purposively selected since it is the biggest institution of higher learning in the county. The inquiry relied on records from the University Admissions Office as its sampling frame. The target population comprised of college and university students in Kenya most of who happened to be of voting age (18 years and above). The units of observation were students in the youth category who doubled up as participants in structured interviews, unstructured interviews, and focus group discussions, whereas the unit of analysis was voter behaviour amongst the youths.

Research Design and Sampling Strategy
This study employed the explorative research design given that it was a fact-finding endeavour aimed at establishing the voter behaviour amongst Kenyan youths. Information obtained from the Admission's Office revealed that students who were in session at the time of the study were those in the Schools of Business, Engineering, Nursing, and Nutrition and Dietetics, whose total was 800. The simple random sampling procedure was applied during the actual data collection. The sample size was determined through the Fisher's formula (Mugenda and Mugenda, 2003).
Where: n is the desired minimum sample size; z is the standard normal deviate at the required confidence level; p is proportion in target population estimated to have qualities being measured; q = 1-p, whereas d is the statistical significance level set n = (1.96) 2 (0.327) (0.673) = 338 respondents (0.05) 2 Those who filled in questionnaires were 338 in number (10 additional respondents were incorporated into the study to cater for possible non-responses). These respondents were picked through the simple random sampling procedure.

Data Collection Instruments, Validity and Reliability, and Data Analysis
Data were collected through observation, structured interviews (questionnaires), unstructured interviews (key informant interviews), and focus group discussions. Before the actual research study was conducted, the researcher administered questionnaires to about 10% of the respondents to gauge the effectiveness of the tool. Construct validity was guaranteed through the researcher's review of relevant literature. Drost (2011) observed that reliability could be improved by crafting items that possess clarity. The study adopted a mix of quantitative and qualitative research strategies. Furthermore, it deployed a descriptive statistical analysis (involving the use of tables) and cross-tabulation between key variables) through the help of Statistical Packages for Social Sciences. Information from focus group participants and key informants was quoted verbatim with the goal of enriching the study. Makori (2015, in a related study, relied upon descriptive statistical analysis of data on a large extent.

RESULTS AND DISCUSSION
3.1 SOCIO-DEMOGRAPHICS OF RESPONDENTS Pintor, Gratschew and Sullivan (2002) observed that pioneer studies on voter behaviour in Western Europe and North America established that gender, age, education and social class were some of the standard sociodemographic characteristics used to predict levels of political behaviour. The socio-demographic characteristics highlighted in this study are: gender, marital status, county of origin, level of education, nature of residence (whether rural versus urban), and religion. Some of these variables were cross-tabulated with voter.

Gender
Data collected through questionnaires revealed that the proportion of males that took part in the research study was higher (64%) than that of females (36%). This is summarized in Figure 1. 100.0 Of the 338 students who participated in the study, the majority (64%) were males whereas the females were in the minority (36%). This distribution mirrored the percentages of the total number of male and female students in the institution (25,631) who were out of session at the time of the study whereby the males were higher (60%) than the females (40%).
The cross-tabulation between gender and party primaries participation showed that males who voted in the 2017 party primaries were 52% (48% failed to) while 54 percent of females participated (46% failed to). The crosstabulation between gender and general election participation indicated that males who indicated they voted in the 2017 general elections, the males who indicated they would vote were 90 percent (10% indicated they would not) while females were 85 percent (15% indicated they would not). The general trend indicated that almost a similar proportion of males and females participated in the 2017 party primaries and a comparable number showed they would take part in the general elections.
In a study conducted in the United States of America, Lopez, Marcelo, Kennedy and Barr (2008) found out that voter registration rates among young men and women were similar in presidential and mid-term election years though there was a small growing gap between young men and women, in relation to the voter turnout rates. Pintor et al (2002) noted that in recent decades the orthodox view that women are less active has been challenged.

Marital Status
Data collected through questionnaires shows that most the university students (84%) are unmarried (92.5%), while 7% are married. This is summarized in Figure 2. The researcher did not seek to find out the relationship between marital status and voter behaviour, and therefore more studies need to be conducted to bring out a clearer relationship between marital status and voter participation. In a study conducted in the United States of America by Lopez et al., (2008), it was discovered that young married people voted at higher rates than their single counterparts did.
Most of these counties are occupied by diverse ethnic communities, and therefore, views from students hailing from most of the tribes in Kenya were captured in this study. Most Kenyans live in geographical locations which reflect their ethnic identity. Kenya's politics are heavily ethnicized given that most people are far more likely to vote for a candidate they share a tribe with.
Makori (2015) observed that majority of the youths voted along tribal lines whereby the interplay of clientelism and strategic and well-calculated political behavior characterized a big portion of their political engagement. Therefore, when a candidate hailing from one's ethnic community is on the ballot, their ethnic identity is enough motivation for the candidate to be voted for by his tribesmen and women.

Nature of Residence
This section was in reference to life in the rural versus life in the urban areas regarding the students. Data collected through questionnaires showed that university students who had lived most of their life in rural zones were 50% while those who spent most of their time in urban areas were 47%. This is presented in Figure 4. The cross-tabulation of nature of residence versus the 2017 general election participation indicated that a slightly higher proportion of students brought up in rural areas (54%) were going to vote (46% indicated they will not) while a relatively lower proportion of those raised up in urban areas (47%) would participate in the general election exercise (53% showed they will not). This goes against the general belief that most young people who have spent a significant amount of time in urban areas were more informed and had a higher probability of involvement in the electoral process. It could also mean that the reasons making those raised in urban areas to keep off the electoral process are different from those in rural areas. However, further studies should be conducted in this regard to gain better insight into the relationship between the nature of residence (whether rural or urban) versus participation in the electoral process.

Religion
Data collected through questionnaires showed that a majority of the students (95.6%) were Christians, 2.7 percent Muslims, while "others" were 0.9 percent. This is illustrated in Figure 5.   Driskell et al (2008) reinforced this by noting that religious beliefs were significantly related to national political participation. A participant in a focus group discussion said: "My religious faith prohibits me from taking part in the electoral process because God would hold me liable for any bad policies enacted by the elected leaders therefore I prefer staying away from the process".

Registration Status for the 2017 General Election
Data collected through questionnaires showed that a majority of the students (91%) had registered for the 2017 general election exercise while 9% had not. This is shown in Figure 9. The overwhelmingly high proportion of voter registration status amongst the respondents could be attributable to voter sensitization drives by Kenya's electoral body. In addition, aspirants for various seats were keen on mobilizing the young people, a key voting bloc, to register in great numbers. Also, a majority of these students are directly affected by socio-economic issues facing the society such as corruption, unemployment, and poor service delivery, and as a result, acknowledge the importance of participation in the electoral process. However, the small percentage that failed to register cited a number of challenges. A participant in the focus group discussions said: "The issuance of national identity cards should be discrimination-free; how fast an ID applicant gets his or her document at times depends on the region they hail from since negative ethnicity is rife in Kenya". Schultz-Herzenberg (2014) observed that while the number of registered voters in South Africa's elections had increased, registration did not always increase at the same rate as population growth rate.

Nature of Reason for the Non-registration
Data collected through questionnaires revealed that more than half (57%) of the 9% of students who failed to register as voters ahead of the 2017 general election exercise gave sound reasons while 43% gave mere excuses. This is presented in Figure 10.  -Herzeberg (2014) pointed out that the gap between the eligible voter population and registered voters called for a reconsideration of the legal requirements for registration processes. The introduction of an automatic registration process, where the state takes the initiative to register eligible citizens, may lessen the costs of registration for some groups. A key official of the Independent Electoral Boundaries Commission buttressed this by saying: "We should consider registering voters the minute they are issued with the national identity card. Also, the services of the electoral body should be decentralized further. Also, the presidential elections, gubernatorial and senatorial elections can be separated from the other contests".

The 2017 Party Primaries' Participation
Data collected through questionnaires indicated that 52% of the respondents participated in the 2017 party primaries whereas 47% failed to take part. One percent did not respond to the question. This is presented in Figure 11. 100.0 A moderate proportion (52%) of the students indicated they would participate in the party primaries, which was an indicator of voter apathy. However, the study probed further into the nature of the reason for failure to take part in the party primaries, as discussed in section 3.1.9.

Nature of Reason for the Non-participation in the 2017 Party Primaries
Data collected through questionnaires indicated that majority (76%) of the 47% students who failed to participate in the 2017 party primaries gave genuine reasons for the failure while 24% of them gave unconvincing reasons. This is summarized in Figure 12. 100.0 Unavoidable reasons for failure to take part in the 2017 party primaries identified by some of the respondents included disorganization and violence at polling stations. A participant in a focus group discussion said: "I think voting in party primaries is a waste of time because most of the party nominees who contest in the general election are pre-determined by party officials". Ignorance, too, is to blame. Wolpe (2007) postulated that one of the key factors that often impeded the effective involvement of Kenyan youth in political processes was ignorance; young voters' lack of understanding of the issues can cause them to become either angry and violent or apathetic and disinterested in the candidates. Schultz- Herzenberg (2014) observed that global studies indicated that young voters were predisposed to lower turnout rates. Caramani (2011) pointed out that the most important institutional regulators of voter turnout were: laws of compulsory voting, electoral rules, registration requirements, and concurrent or non-concurrent elections for legislative and presidential office.

2017 General Election Participation
Data collected through questionnaires showed that 88% of the students indicated they will participate in the 2017 general elections while 10% signaled they would not be able to take part. Two percent were unsure. This is summarized in Figure 13. However, Caramani (2011) observed that generally poorer and more authoritarian countries tended to record much lower voter turnouts, but Kenya has made significant strides in matters to do with democracy.
3.1.11 Nature of Reason for Non-participation in the 2017 General Election Data collected through questionnaires indicated that slightly more than half (58%) of the 10% of students who indicated they would fail to take part in the general election gave convincing reasons while 42% gave feeble reasons. This is illustrated in Figure 14.  Wolpe (2007) opined that a key factor which impeded the effective involvement of youth in general elections was ignorance. Youthful voters' lack of understanding of the issues affecting the country caused some of them to become both angry and violent or apathetic and disinterested in the candidates. This point is reinforced by a focus group discussion participant who said: "Voting in elections will neither get my bills paid nor bring food to my table". One gave good suggestions for consideration: "Most of us wish to participate in the general election exercises but it becomes an uphill task when we are forced to travel to our home counties hundreds of kilometers away to do so; students in colleges and universities should be able to register as voters from any part of the country. As for the polling stations used during general elections, they should be as close as possible to the voter".

Importance of Voting in General Elections
Data collected through questionnaires indicated that majority of the respondents (97%) acknowledged that voting in general elections in Kenya was important while 2 percent did not see the importance. One percent failed to respond to the question. This is presented in Figure 15. 100.0 Voter participation, especially by the youths who form a significant proportion of the population in Kenya, injects political legitimacy into the electoral process. Schultz-Herzenberg (2014) reinforced this by noting that the steady decline in electoral participation since 1994 may weaken the legitimacy that is the lifeblood of democratic institutions and eventually erode the overall quality of the democracy in South Africa. Ntau and Ntsabane (2000) buttressed this further by observing that one of the vital ways of ensuring the survival of any electoral system was not only in the application of the law to the letter but also in enlisting the development of a democratic culture, and that in the event that a significant part of the population loses confidence in the electoral system, then the rulers lose their legitimacy. The negative trend of election rigging especially at the presidential level was cited by most respondents and participants in a focus group discussion as one of the key reasons they lost faith in the power of the ballot. One of them lamented: "General elections in Kenya are never free, fair and transparent". Tracey (2016), in a similar study in South Africa, pointed out that young people generally perceived politics as being characterized by rampant corruption, and that they referred to it as a dirty game, a waste of time, and as an endeavour they would not wish to get involved in. To them, bad leadership, corruption, and fraud by "criminals in Parliament" were some of the issues which deterred them from active participation in politics. She concluded that the participation of young people in the electoral process was often dependent on their satisfaction with democracy and the performance of the ruling party and its political leaders in addressing the issues which directly affected them. In this research study, the majority of university students (97%) expressed confidence in voting as a democratic process, which is a positive sign. Lopez et al (2008) pointed out that empirical evidence suggested that once someone voted, returning to vote a second time was easier, and could lead to a lifetime habit of voting. However, with the claims of massive rigging of the election outcome and the subsequent nullification of the August 8, 2017 presidential election results (please refer to the Judgment by the Supreme Court of Kenya in the Presidential Petition 1 of 2017) that ushered in a repeat presidential election which was boycotted by the opposition coalition, the National Super Alliance (NASA), it remains to be seen if the level of optimism in the electoral process will be adversely affected in the next electoral cycle five years down the line.

CONCLUSION
The study revealed that 52% of respondents had taken part in the 2017 party primaries, 47% did not, while 1% failed to give a clear response. Of the 47% who did not take part in the 2017 party primaries, the majority (76%) gave valid reasons whereas the remainder (24%) gave unconvincing reasons. Eighty-eight percent of respondents indicated they would participate in the 2017 general elections, 10% would be unable to, while 2% were unsure. Of the 10% who were not going to vote, 58% of them had valid reasons for their failure to do so whereas 42% gave unconvincing reasons. In relation to registration for participation in the 2017 general elections exercise, an impressive 91% had registered to vote while only 9% had not. On the importance of voting in general elections, an overwhelming majority of 97% argued in favour of participation, 2% against, while 1% failed to give a clear response.
Therefore, this inquiry concluded that despite hurdles discussed herein facing Kenyan youth regarding participation in the electoral process, most of them recognized the importance of voting in party primaries and general elections, and generally manifested a strong positive attitude towards voting. However, as noted earlier, following the nullification of the 2017 presidential election and eventual boycott of the fresh presidential election by the major opposition coalition, the National Super Alliance (NASA), it is quite likely that a significant proportion of Kenyans lost confidence in the power of the ballot.

RECOMMENDATIONS
Since perceptions are dynamic, the researcher recommends the execution of a similar study before Kenya's next election cycle in 2022. Also, the legislative, judicial and executive arms of government should consider making voting in the general elections mandatory while simultaneously instituting radical electoral reforms that will make the process easier. Furthermore, the playing field should be made level by ensuring the winner of any political contest is determined by voters and not by those counting the votes.