FACTORS AFFECTING MILLENNIAL GENERATION VOTING DECISION IN THE 2019 GENERAL ELECTION IN SURABAYA

An election in a democratic country has become routine in determining the regeneration of leadership. Political participation, especially voting in elections, is the key to democratic governance. Millennial Generation or Millennials have great potential for political power because of their large numbers. Their voices cannot be taken lightly. Instead, Millennials can make a significant vote contribution for the presidential candidates. The formulation of the problems discussed in this study is about how leadership, emotional relations, mass media, political issues, and voters' attitudes affect the decision to elect Presidential candidates and Legislative candidates by millennial generations. This study was a quantitative study in which the data were collected by distributing questionnaires to millennials who lived in Surabaya City. The obtained data were then analyzed using Partial Least Square (PLS) analysis method. The findings showed that leadership, emotional connection, mass media, and voter attitudes had a positive significant effect on choosing candidates by the millennial generation. Meanwhile, political issues did not significantly influence their decision in choosing candidates in the 2019 general election in Surabaya.


INTRODUCTION
The general election is a mean to determine the direction of the nation's journey as well as to determine the most suitable leader to govern the country (Sardini 2011:298). The general election in a democratic country has been routine in determining the regeneration of leadership. Political participation, especially voting in an election, is the key to democratic governance. During this election moment, the public can participate in determining their leaders. Commissioner of the General Election Commission, Viryan Azis, said that voter participation in the 2019 election increased compared to the previous election. Compared to the 2014 election, the increase in participation rates was nearly 10 percent. The number of public participation in the 2019 election was 81 percent. It was increasing compared to the 2014 presidential election which was only 70 percent and the 2014 legislative election which was 75% (Farisa, 2019). KPU believed the number of public participation in the 2019 election was 81%. This was said to have exceeded the target set by the KPU which was only 77.5% (Andayani, 2019). The increase in national-level voter participation was contributed by one of the city with the largest population in Indonesia, Surabaya. The level of community participation in Surabaya has also increased. Based on the data of Surabaya KPU, during the 2014 election, the number of voters in Surabaya City was 2,017,450 voters. The number increased in the 2019 election which became 2,131,756 voters (KPU, 2019). This supported the results of the Open Plenary Meeting Recapitulation and Determination of the Third Revised Permanent Voters List (DPTHP-3). DPTHP-3 determined the number of voters in Surabaya City was 2,131,756 voters (Hakim, 2019).
Hence, this study focused on voters, specifically the millennial generation with an age range of 17 to 35 years old. Millennials have great potential for political power because of their large numbers. Millennial generation votes cannot be taken lightly. Instead, their votes can make a significant vote contribution for the presidential candidates. Voters which were the millennial generation were also known as the generation of technology and information literacy. They were active on social media. Accordingly, they can easily access the profiles and track records of presidential and vice-presidential candidates. They were also active in online discussions through social media. It cannot be denied, due to some reasons, the millennial generation was torn apart between those who supported the presidential and vice-presidential candidate pair number 01, Joko Widodo and Ma'ruf Amin, and those who supported the presidential and vice-presidential candidate pair number 02, Prabowo Subianto and Sandiaga Uno. Some of them even established groups or volunteers supporting the presidential and vicepresidential candidates (Apandi, 2019). The candidates and political parties who contested in the 2019 Presidential Election were obliged to accommodate the millennial generation in policy and to prepare a better political future. The 2019 legislative and presidential election needed to show further evidence of how millennial has been transformed as an important political subject not just a statistic of voters. There may be hope for a renewal of political future in the future (Irawan, 2018).
Based on the background description above, it can be arranged several formulations of the problem to be investigated, including: 1). How does the leadership influence on the decision to choose the candidates for President candidates and Legislative candidates by millennial generation; 2). How does the emotional relationship affect the decision to choose the Presidential candidates and Legislative candidates by millennial generation; 3). How is the influence of the mass media on the decision to choose candidates for President and Legislative candidates by millennial generations; 4). How the influence of political issues on the decision to choose candidates for President and Legislative candidates by millennial generation; and 5). How is the influence of voters' attitudes on the decision to choose candidates for President candidates and legislative candidates by millennial generation. In previous studies, the object of the study was all voters who were not limited to a certain age. Meanwhile, this study focused on millennial generation voters. This study aimed to analyze the factors that influence voters from the millennial generation in choosing candidates. This study was expected to provide appropriate input and suggestions for candidates to formulate political strategies and to find out what voters consider in deciding on choosing candidates.

General Election
General Election is a means to realize the sovereignty of the Indonesian who adheres to a democratic political system. This is stated in Law No. 7 of 2017 concerning General Election which explains that the General Election is held on the basis of direct, general, free, confidential, honestly, and fair once in five years to elect the President and Vice President as well as members of the DPR (the house of representatives), DPD (Regional House of Representative), and DPRD (Regional House of People's Representatives) members. The holding of presidential election shall be held to elect a president and vice president who have strong support from the people. Thus, they can carry out the functions of the power of the state government to achieve national goals as mandated in the opening of the 1945 Constitution of the Republic of Indonesia.
The election is one of the main pillars of an accumulation of people's will as well as democratic procedures for electing leaders in a democratic country. Through the General Election, the people elect their representatives. After that, the elected people's representatives are entrusted with people's sovereignty mandate to take care of the country. The election is the main mean of realizing democracy in a country. The substance of the election is the submission of the people's voice to form representative institutions and the government as the organizer of the state. The people's voice is manifested in the form of voting rights which is the right to vote for candidates to represent them in government.
According to Law No. 10 of 2008 concerning the General Election of the DPR, DPD, and DPRD members, the principle of general election includes: a. Direct, it means that voters have the right to directly vote according to their conscience without intermediaries. b. General, it means that all citizens who are 17 years old or married have the right to vote.
Additionally, citizens who are 21 years old or above have the right to be chosen with no discrimination (exceptions). c. Free, it means that voters have the right to vote according to their conscience without any influence, pressure, or coercion from anyone / with anything. d. Confidential, it means that the voters are guaranteed by regulations will not be known by anyone and in any way who is chosen or to whom the vote is given (secret ballot). e. Honestly, in organizing an election, organizers, government, and political parties participating in the election, supervisors and election observers, voters, as well as parties involved indirectly are obliged to be honest following applicable laws and regulations. f. Fair, in holding every election, political parties participating in the election receive the same treatment and are free from cheating by any party.

Decision to Vote Candidates
Voter's decision to determine their political choice is based on three things, namely problem solving (fulfillment of needs), rational (function and usefulness), and hedonic benefits (emotions, tastes, and aesthetics). Other factors that might influence voter behavior are the cultural environment, social class, personal and family influence, and the local situation (Tabroni, 2014:126). Choosing candidates in the general election is almost the same as choosing a product to buy. Cwalina et al. (2008) emphasized that by looking at the problem from a consumer perspective, a broader marketing approach can make a useful contribution to better theoretical knowledge of the "decision-making process." The core of the buying process involves the following stages: a. Problem Recognition: This stage refers to motivation which triggers recognition that there are issues to be considered. In essence, the process boils down to the question to voters: "Whom will I vote for?" Recognition of the problem is determined by the needs of the voters, who, at different levels, refer to the problem of voting for candidates. For example, if a voter has problems getting a job, he will be sensitive to programs where politicians emphasize lower unemployment as one of their main objectives. b. Information Search: Voters seek the various source of information from TV, radio, newspapers, magazine, et cetera which highlights recognition of problems. Naturally, each source may have a different effect on the voters' opinion. The main sources of consumer information are divided into four groups: a) Personal such as family, friends, neighbor, colleagues; b) Commercial such as advertising, website, salesforce, distributor, packaging, display; c) Public such as mass media, organizations, consumer ratings, d) Experimental such as handling, checking, product usage. c. Alternative Evaluation: Voters need to consider the obtained information to a set of evaluative criteria. These criteria are related to voter motivation which refers to the first stage of the decision-making process which is problem recognition. If voter evaluative criteria are very under their motivation, then it is very difficult to cause changes in their voting behavior. d. Decision Selection: Choice is a very important element of the decision-making process. It needs to be logically connected with alternative evaluations. However, voters can change it at the last moment due to the influence of the last minute like an article that is read, a new broadcast watched, or a debate with a friend. These unexpected situational factors are mainly related to last-minute voters including those in the undecided voters. e. Post-Election Behavior/Results: This element corresponds exactly with post-purchase behavior in consumer behavior. A politician needs to maximize voter satisfaction, including those who do not vote for him. Ongoing community relations activities and the political patronage of influential groups can achieve this goal.

Factors That Influence Election Decisions
Several factors that influence the decision to vote in an election can be described through the following explanation: a. Leadership. Daft in Munandar (2014) defined leadership as a related influence between leaders and followers who want change and a tangible outcome that reflects the common goal. Alie (2013:68) stated that a candidate's leadership consists of credibility (quality, capability or strength to generate trust), ability (in the economic field, security, and welfare of the people), and personality (having the personal character of a leader. Leadership influences voting behavior, where voting behavior is an activity of voting by individuals that is closely related to the decision-making activities to vote and not vote in an election, voters will vote or support certain candidates. (Mustanir & Jaya, 2016) b. Emotional connection. Cwalina et al. (2008) conducted a further study on voting behavior using Newman and Sheth's theory. The findings concluded that emotional connection is one of the independent elements and also a giant predictor that influences the voting decision. Alie (2013:72) said that an emotional connection is a judgment that can influence the voting decision. According to Alie (2013:75), emotional connection consists of similarity in religion, organization, ideology, and ethnicity between candidates and voters. The level of political participation in the community is very dynamic, but the problem is related to motivation, most people admit that they choose to be driven by transaction factors and emotional closeness. The candidate's vision and mission are not a measure of choice. (Liando, 2016) c. Mass Media. Firmanzah (2012:54) defined mass media as a factor that can influence voters in choosing political candidates. The mass media consists of exposure from experts, political advertisements, seminar results, surveys, political news, and so on which are disseminated to the public. Winchester (2014) defined mass media as a source of political information that influences voter behavior before the election. Information from the mass media can come from television, Internet, social media, radio, billboards, and so on. Political advertising in the post-conflict local election has a strong influence on the participation of election voters. Participation in this matter is the participation of a person in the election of state leaders, both directly and indirectly. If it is associated with the theory of moderate effects which assumes that the influence of the mass media is not in an unlimited or limited position, it will depend very much on the individual who is exposed to the mass media message. (Fenyapwain, 2013) d. Political Issues. Political issues refer to voters' personal beliefs about issues that are broadly addressed to candidates in the political world including economic, social, foreign influence, supporters, and leadership characteristics (LSI Denny JA, 2018). Yunita dan Stainlaus (2014) mentioned that cognitive orientation which consisted of aspects of political issues and policies was not the basis of voters in choosing a legislative candidate. e. Attitude. Someone's attitude toward a behavior is defined as one's evaluation of the behavior which is based on the expected positive and negative results related to belief behavior (Ajzen 2015). Kotler and Keller (2016:197) defined attitude as an evaluation, emotional feeling, and tendency of actions that someone likes or dislikes toward an object or idea. Although many definitions of attitude have been put forward, all these definitions have a common similarity. Attitude is interpreted as an evaluation of someone (Sumarwan 2011:165). Retnawati et al. (2015) states that there is an influence of political attitudes on student political participation in elections, which means that the higher the political attitude, the higher the political participation in elections. Based on some of the above meanings, the conceptual framework in this study was designed as follows:

RESEARCH METHOD
This study was quantitative research to test the hypotheses that had been set. The variables in this study were divided into two which were exogenous variables (leadership, emotional relations, mass media, political issues, and voter attitudes) and endogenous variable (the decision to vote for a candidate  Cwalina et al. (2008) This study employed a survey method to collect the data in which a critical direct observation or investigation to get clear information from the source for a particular problem in an area. The main instrument used was a list of questions (questionnaire) using a Likert scale.
The population of this study was all voters including millennial generation in Surabaya. This study involved 100 people who were in millennial generation domicile in Surabaya as the sample. The obtained data were analyzed using Partial Least Square (PLS) analysis method.

Respondent Characteristics
The characteristics of the respondents in this study who were a millennial generation in the 2019 legislative and presidential election in Surabaya were reviewed based on their address, gender, age, educational background, and participation frequency in the election. The results were presented as follows:  Table 2 presents the characteristics of the voters who were the most dominant research respondents. In this table, it is known that the majority of voters reside in North Surabaya, 37 people (37%). Based on the gender, the majority of voters in the 2019 legislative and presidential election were male (62%). Meanwhile, the majority of voters who were at the age of 21 -35 years were 74 people (74%). the majority of voters in general completed high school/equivalent level of 46 people (46%). By the research topic, the final characteristics were the frequency of participation in elections. There were 71 people (71%) who had participated in the election more than twice. The remaining, 29 people (29%), had participated in the election once.

Outer Model
Evaluation on the outer model consisted of two kinds namely evaluation on reflective indicators and evaluation on formative indicators. Measurement models (outer models) regarding the influence of leadership, emotional connection, mass media, and political issues on voter attitudes and the effect of voter attitudes on the decision to vote for a candidate were visualized as follows:

. Outer Model
From the measurement model (outer model) in Figure 2 above, the data were evaluated with the following criteria: 1) Reflective Indicator a. Validity test Figure 2 above showed that the outer loading value of the indicators of the latent constructs of leadership, emotional connection, mass media, political issues, voter attitudes, and the decision to vote for a candidate was more than 0.7. Therefore, it can be concluded that the measurement model formed had convergent good validity. Outer loading value on indicators with latent constructs was higher compared to its cross-loading value. Accordingly, the measurement model had good discriminant validity. Therefore, indicators that measured the latent constructs of leadership, emotional connection, mass media, political issues, voter attitudes, and decided to vote for the candidate were valid. b. Reliability Test Reliability test was done by looking at the composite reliability value of the indicator block that measures the construct. The reliability composite results showed a satisfactory value above 0.7. 2) Formative Indicators a. Outer Weight The formative indicator weight value with the construct has to be significant. If the value of t-statistic outer weight >1.96, it means the value of formative indicator weight with a significant construct. The significance of the indicator with its construct was visualized in Figure 3 of the inner model. In the figure, t-statistics obtained more than 1.96 values. Therefore, the value of weight indicators that measures each construct of leadership, emotional connection, mass media, political issues, voter attitudes, and the decision to vote for a significant candidate. b. Multicollinearity VIF value > 10 means that the indicator did not show multicollinearity. The VIF values for each indicator were as follows:  Table 4 above explained that the VIF value on each indicator was less than 10. Thus, it can be concluded that there was no multicollinearity in the indicators of the constructs of leadership, emotional connection, mass media, political issues, voter attitudes, and the decision to vote for candidates.

Inner Model
Structural model (inner model) regarding the influence of leadership, emotional relations, mass, media, political issues, and voter attitudes towards the decision to vote for a candidate was as follows: Jurnal Politico Vol.19 No.2 September 2019 Halaman 87-102. ISSN: p;1829-6696, e:2549 Web jurnal online; jurnal.unmuhjember.ac.id Evaluation on the inner model consists of two types namely the value of R-Square (R2) and hypothesis testing with the description as follows: 1) R-Square (R 2 ) Value R-Square value on the influence path of leadership, emotional connection, mass media, political issues, voter attitudes toward the decision to vote for a candidate was obtained as much as 46.0%. Meanwhile, the remaining 54.0% was explained by other variables not examined.

2) Hypothesis Testing
Exogenous variables are said to have a significant effect on an endogenous variable if the tstatistics > 1.96 (normal Z-score). The results of the hypothesis testing were as follows: obtaining positive value with t-statistics values of 2.971, 2.628, 2.108 and 2.155 more than 1.96. Thus, leadership, emotional connection, mass media, and the attitude of voters had a positive and significant influence on the decision to vote for a candidate. It shows that the leadership of the candidate, the emotional connection between voter and the candidate, the mass media related to the candidate and the better the attitude of the voter, the higher the probability of the candidate to be chosen by the voter. Then the path coefficient of influence of political issues on the selection of candidate was negative with the t-statistics value of 1.367 less than 1.96. Therefore, political issues did not significantly influence the decision to vote for a candidate. In other words, political issues do not necessarily affect the decision of voters to vote for a certain candidate because the political issues depend on the viewpoints of each voter in responding to them. The voters referred to here are voters from the millennial generation. This generation is classified as a generation of technological literacy. Hence, they did not take political issues in societies for granted. They always look for lots of information to find the truth of the issue.

Discussion
Judging from the results of the analysis above, it can be seen that leadership had a positive and significant effect on the decision to vote for a candidate. Accordingly, the hypothesis that leadership influences the decision to vote for a candidate in the 2019 election in Surabaya was proven to be true. This indicates that if the candidate has good leadership, then the possibility of a candidate being chosen by voters will get higher. According to Daft in Munandar (2014), leadership is a related influence between leaders and followers who want change and a tangible outcome that reflects a common goal. Alie (2013:68) revealed that a candidate's leadership consists of credibility (quality, capability, or strength to generate trust), ability (in the economic field, security, and welfare of the people), and personality (having the personal character of a leader). The finding of this study supports the empirical study carried out by Widagdo et al. (2014). The political products consist of political parties, previous track records, and personal characteristics are factors that influence behavior in selecting candidates. The finding of this study also supports an empirical study conducted by Mustanir and Jaya (2016) which concluded that leadership influences voting behavior. Voting behavior is an activity of voting by individuals that is closely related to the decision-making activities to vote or abstaining in a general election.
The emotional connection had a positive and significant influence on the decision to vote for a candidate. Hence, the hypothesis stating that an emotional connection influences the decision to vote for a candidate in the 2019 election in Surabaya was proven to be true. In other words, if there is an emotional connection between the voter and the candidate, the possibility of the candidate being chosen by the voter will get even higher. Cwalina et al. (2008) stated that emotional connection is the dominant factor that can influence the voting decision. Alie (2013:72) said that emotional connection is a judgment about the emotional electorate of parties or candidate that can influence voting decisions. Alie (2013:75) defined emotional connection as a judgment about the emotional electorate of parties or candidates that can influence voting decisions. According to Alie (2013), emotional connection consists of similarity in religion, organization, ideology, and ethnicity between candidates and voters. The findings of this study were following the empirical study conducted by Daud M. Liando (2016) where the level of political participation of the community is very dynamic. However, the problem was related to motivation. Most people acknowledged that they vote due to the transaction factors and closeness emotional elements. The candidate's vision and mission were not their measures in voting. Another empirical study by Fernandes (2018) stated than in the Jakarta case, the importance of identity (ethnic and religious) factors in voter behavior can be seen in studies conducted by Prasetyawan (2014) and Gani (2018). Their study showed that ethnicity still plays an important role in affect voter behavior. Prasetyawan focused more on the ethnic factor of the candidate while Gani (2018) showed that the identity factor, religion, is one of the determinants of the candidate's victory. The results of this study contradicted with the findings of Haryanto's (2014) study which concluded that the emotional closeness between voters and the prospective candidate was not so dominant in determining voting behavior in East Luwu.
The mass media had a positive and significant influence on the decision to vote for a candidate. Nevertheless, the hypothesis stating that the mass media affects the decision to vote for a candidate in the 2019 election in Surabaya was proven to be true. This finding supported the finding of an empirical study conducted by Gabriela Ribes-Giner and Maria Fuentes-Blasco (2013). They concluded that political marketing is centered on candidates with the biggest influencing factor being through the media. An empirical study conducted by Febyapwain (2013) also concluded that political advertising in the post-conflict regional election has a relatively strong influence on voter turnout participation. Participation in this matter is the participation of a person in the election of state leaders, both directly and indirectly. If it is associated with a theory of moderate effects which assumes that the influence of the mass media is not in an unlimited or limited position, it will depend very much on the individual who is exposed to the mass media message. Another empirical study by Retnawati et al. (2015) also stated that there is an influence of mass media on students' political participation in elections. Therefore, the more political news in the mass media, political participation will get higher. However, the finding of this study contradicts the finding of Addasuqi's (2015) research which concluded that there was no influence between political advertising and decision making.
Political issues did not significantly influence the decision to vote for a candidate. Thus the hypothesis stating that political issues affect the decision to vote for a candidate in the 2019 election in Surabaya was not proven true. The result of this study supported an empirical study conducted by Yunita and Stainlaus (2014) which concluded that in a cognitive orientation which consisted of political issues and policies aspects was not the basis of voters in voting for a legislative candidate. This can be seen on the results of the study which showed that in the aspect of vision, mission, and work programs, the majority of respondents (54.8%) were in the low category with the number of respondents in the high category of 11.5%. Low category means that voters did not have the knowledge and trust in the policy issues as well as vision and mission presented by legislative candidates during the campaign period. Yet, this is not consistent with the results of empirical study by Axhami et al. (2015) which concluded that political marketing was closely related to political campaigns in which the main elements in this political campaign were party programs related to economic and social issues, national issues, leader personality, party image, and party ideology. To increase voter support, political parties were required to focus on those elements. Additionally, another empirical study by Peranginangin and Zainal (2018) which stated that voters seek political information related to current issues. This information serves as a reference in making decisions in choosing a president and vice president at the general election.
The attitude of voters had a positive and significant influence on the decision to vote for a candidate. Thus, the hypothesis stating that the attitude of the voters affects the decision to vote for a candidate in the 2019 election in Surabaya was proven to be true. This indicates that the better the attitude of voters towards candidates, the decision of voters to vote for the candidate will be better. In other words, the higher the possibility of a candidate to be chosen by voters. This result supported the results of a study by Hilda Coffe and Elizabeth Theirss-Morse (2016) stating that voters determine candidates based on the most competent political experience in dealing with the most competent business backgrounds in handling economic problems. Also, an empirical study by Retnawati et al. (2015) also stated that there is an influence of political attitudes on student political participation in elections. It means that the higher the political attitude, the higher the political participation in elections. Jurnal Politico Vol.19 No.2 September 2019 Halaman 87-102. ISSN: p;1829-6696, e:2549 Web jurnal online; jurnal.unmuhjember.ac.id By: Citra Oktavia Mochtar Factors Affecting Millennial Generation Voting Decision In The 2019 General Election In Surabaya 100 CONCLUSION Based on the previously described results of the analysis and discussion, it can be concluded that leadership, emotional connection, mass media, and the attitude of voters from the millennial generation in Surabaya have a significant influence on the attitude of voters in the 2019 election n Surabaya. This means that if the candidate's leadership is good, there is an emotional connection between the voter and the candidate, the mass media is increasingly playing an active role and the better the attitude of the voter, the decision of the candidate voted by the voter will be higher. Meanwhile, political issues did not significantly influence the decision to vote for a candidate. Leadership, emotional connection, mass media, and the attitude of voters have an important role in making a voting decision. Accordingly, candidates need to focus on these four factors to gain more voter support, especially from the millennial generation.