Eduvest – Journal of Universal Studies Volume 3 Number 4, April, 2023 p- ISSN 2775-3735- e-ISSN 2775-3727

THE IMPACT OF POLITICAL MARKETING MIX ON REPEATED VOTING DECISION IN INDONESIAN GENERAL ELECTIONS: A CASE STUDY OF FEB UNIVERSITAS RIAU STUDENTS
Alvi Furwanti Alwie¹, Rendra Wasnury², Jushermi³, Nia Anggraini⁴^, Monica Agnes Br Harianja⁵, Dedi Hidayat⁶ Faculty of Economics and Business, Universitas Riau, Indonesia¹²³⁵⁶ STIE Persada Bunda, Indonesia⁴ Email: [email protected], [email protected], [email protected], [email protected], [email protected], [email protected]
ABSTRACT
Politics is the basic policy of the state administration that is currently and will apply, which is sourced from the values that apply in society to achieve state goals. The purpose of government-run politics is to achieve the welfare of the general public. Elections are a form of political participation as an embodiment of people's sovereignty, because at the time of voting, the people become the most decisive party for the political process in an area by voting directly. This study aims to determine whether the influence of the Political Marketing Mix on the decision to choose repeat voters in General Elections in Indonesia (a case study on FEB students, Universitas Riau). This study applied quantitative research method. The analysis technique used is multiple linear regression with the test equipment using Warp PLS 7.0. The number of samples to be taken as many as 145 respondents with purposive sampling method. The data was collected by distributing online questionnaires. Based on the results of the study, it was found that: 1) Political products have a significant effect on voting decisions. 2) Political Promotion has a significant effect on the Decision to Choose. 3) Political Price has a significant effect on the Decision to Choose. 4) Political distribution has a significant effect on voting decisions.
KEYWORDS	political marketing mix; voting decisions; repeat voter
	*This work is licensed under a Creative Commons Attribution-ShareAlike 4.0 International*

INTRODUCTION

Politics is the basic policy of administering the state which is currently and will apply, which originates from the values prevailing in society to achieve state goals (Pramono, 2018; Qurbani, 2012). The purpose of government-run politics is to achieve the welfare of the general public. Presidential elections are a form of political participation as a manifestation of people's sovereignty, because at the time of voting, the people become the most decisive party to the political process in an area by voting directly (Filan & Firdaus, 2022).

Citizens' political awareness is an important factor in political participation, meaning that it is related to knowledge and awareness of rights and obligations related to the community environment and political activities being the measure and level of someone involved in the process of political participation (Averus & Alfina, 2020; Nanda, 2017). The experience of general elections in Indonesia which have taken place over several decades shows that many voters do not vote (Ta’dung, 2017). As a phenomenon described above, if someone has high political awareness and trust in the government, then political participation tends to be active, whereas if awareness and trust are very small, then political participation becomes passive and apathetic (Saputra, 2017).

Data from We The Youth shows that in the 2014 election there were 24.89 percent of voters who decided to abstain (Shofi et al., 2020). In the 2019 election, the General Election Commission recorded the number of permanent voters under the age of 20 as many as 17.5 million people, and voters aged 21 to 30 as many as 42.84 million people. In total, the number of voters aged less than 20 years and 21-30 years reached 60.34 million people. The total number of permanent voters for the 2019 election is 192 million. This means that 31.4 percent of voters in the 2019 election are young people.

Table 1. Number of Voters by Age in the 2019 Election

No	Age	Total
1.	17-20	17,5 million people
2.	20-30	42,84 million people
3.	30 (above)	132,46 million people
Total 192,8 million people

Source: Tempo.co

The Alvara Research Center survey also explains why young people are the biggest contributors to abstentions (Silalahi, 2019). First, it could be because they are not very active in following political news. Second, they tend to be apathetic towards political processes. Third, their knowledge related to politics is not too great. More specifically, based on data from the Riau Provincial Election Commission in 2019, it shows the high number of abstentions in Riau. This can be seen from the data on voters in the DPT who do not exercise their right to vote. The following details the data that the author presents.

Table 2. Data on the number of voters and abstentions from the 2019 Riau Province election

No	Descriptions	Details
A	Voter Data	Voter	Voting Rights	Abstentions	Percentage
1	Pekanbaru City	611.093	482.116	128.977	21%
2	Kampar	510.475	412.067	98.408	19%
3	Meranti Island	147.517	106.181	41.336	28%
4	Indragiri Hulu	301.342	239.190	62.152	21%
5	Bengkalis	412.262	321.644	90.618	22%
6	Indragiri Hilir	491.150	347.179	143.971	29%
7	Pelalawan	226.417	183.598	42.819	19%
8	Rokan Hulu	339.328	276.455	62.873	19%
9	Rokan Hilir	417.327	318.531	98.796	24%
10	Siak	297.161	233.809	63.352	21%
11	Kuantan Singingi	232.018	190.142	41.876	18%
12	Kota Dumai	199.963	160.168	39.345	20%
	Total	4.186.053	3.271.530	914.523	22%

Source: KPU Riau Province, 2019

The high potential of voters among young people, especially repeat voters with the dark side of indifference to elections, creates a big challenge and homework for political parties and candidates to convince and get young people to wake up to vote. This makes the candidates must have a pattern to market themselves. Moreover, this study aims to determine whether the influence of the Political Marketing Mix on the decision to choose repeat voters in General Elections in Indonesia (a case study on FEB students, Universitas Riau).

RESEARCH METHOD

Research sites

This quantitative research was conducted at the Faculty of Economics and Business, Universitas Riau.

Population and Sample

In this study, the population was all active FEB students at the Universitas Riau. The sample in this study used a nonprobability sampling method with a purposive sampling method, namely sampling based on certain criteria.

Determination of the number of Representative samples depends on the number. The number of indicators is multiplied by 5 to 10 (Ferdinand, 2014). So the number of representative samples in this study is: 29x5=145. The criteria set by the researcher in the sampling study are:

1) Active student of FEB, Universitas Riau

2) Have been a repeat voter (using their right to vote at least 2 times to elect both the President and Vice President, DPR/DPRD, Regent and/or Mayor election.

Data analysis

The data analysis used is Multiple Regression Analysis using WarpPLS version 7.0 software. PLS analysis has two models, namely the outer model and the inner model. The outer model (measurement model) specifies the relationship between variables and their indicators. Meanwhile, the inner model (structural model) specifies the relationship between latent variables, namely between exogenous/independent variables and endogenous/ dependent variables (Ghozali, 2008).

RESULT AND DISCUSSION

Data Quality Test (Outer Model) Convergent Validity Test

Table 3. Data Quality Test (Outer Model) Convergent Validity Test

Variable	Indicator	Loading	Decision	AVE
Product	PDP1	0,753	Valid	0,617
	PDP2	0,811	Valid
	PDP3	0,758	Valid
	PDP4	0,754	Valid
	PDP5	0,755	Valid
	PDP6	0,872	Valid
	PDP7	0,853	Valid
	PDP8	0,785	Valid
	PDP9	0,713	Valid
Promotion	PMP1	0,781	Valid	0,568
	PMP2	0,79	Valid
	PMP3	0,806	Valid
	PMP4	0,812	Valid
	PMP5	0,762	Valid
	PMP6	0,752	Valid
	PMP7	0,749	Valid
	PMP8	0,749	Valid
	PMP9	0,723	Valid
	PMP10	0,707	Valid
	PMP11	0,746	Valid
	PMP12	0,734	Valid
	PMP13	0,708	Valid
	PMP14	0,718	Valid
Value	HP1	0,788	Valid	0,655
	HP2	0,873	Valid
	HP3	0,763	Valid
Distribution	DP1	0,744	Valid	0,702
	DP2	0,877	Valid
	DP3	0,886	Valid
Decision	KM1	0,782	Valid	0,714
	KM2	0,77	Valid
	KM3	0,912	Valid
	KM4	0,907	Valid

Source: Research Processed Results, 2021

In the table above it can be seen the value of the loading indicator or loading factor construct of each variable. It is known that all indicators have a loading factor value above 0.7. Then also obtained an average variance extracted (AVE) value above 0.50 which means that all the reflective indicators above have a correlation with the construct variable. This explains that all indicators in the variable construct meet the convergent validity requirements.

Discriminant Validity Test

Table 4. Results of Discriminant Cross Loading Validity Test of Research Variables

Indicator	Product	Promotion	Value	Distribution	Decision
PDP1	0,753	0,097	-0,064	0,059	0,041
PDP2	0,811	-0,104	0,083	-0,172	0,201
PDP3	0,758	0,075	0,075	0,012	-0,068
PDP4	0,754	-0,048	0,115	0,078	-0,25
PDP5	0,755	-0,007	-0,105	-0,13	0,227
PDP6	0,872	-0,06	-0,117	0,01	-0,019
PDP7	0,853	-0,029	0,059	0,02	-0,08
PDP8	0,785	0,055	-0,088	0,031	0,045
PDP9	0,713	0,042	0,051	0,105	-0,106
PMP1	0,045	0,781	0,242	0,149	-0,432
PMP2	0,026	0,79	0,314	0,125	-0,396
PMP3	-0,03	0,806	0,279	0,032	-0,257
PMP4	-0,117	0,812	0,303	0,049	-0,209
PMP5	0,114	0,762	-0,032	-0,006	0,094
PMP6	0,184	0,752	-0,008	-0,145	-0,05
PMP7	0,014	0,749	-0,017	-0,059	-0,024
PMP8	-0,016	0,749	-0,038	-0,1	-0,026
PMP9	0,022	0,723	0,095	-0,173	0,214
PMP10	-0,176	0,707	0,158	0,171	-0,271
PMP11	0,016	0,746	-0,404	-0,032	0,376
PMP12	-0,067	0,734	-0,353	-0,014	0,375
PMP13	-0,024	0,708	-0,389	-0,004	0,468
PMP14	0,005	0,718	-0,251	-0,008	0,253
HGP1	0,154	-0,062	0,788	-0,144	-0,143
HGP2	0,009	0,097	0,873	0,069	-0,055
HGP3	-0,168	-0,047	0,763	0,07	0,211
DBP1	0,137	0,028	-0,099	0,744	0,02
DBP2	-0,087	-0,028	-0,014	0,877	0,041
DBP3	-0,029	0,004	0,096	0,886	-0,057
KM1	0,103	0,058	-0,172	0,161	0,782
KM2	-0,077	0,195	0,082	0,189	0,77
KM3	-0,014	-0,125	0,038	-0,15	0,912
KM4	-0,009	-0,09	0,04	-0,149	0,907

Source: Research Processed Results, 2021

In the table above it can be seen that the correlation value of all indicators from each construct has a high correlation with the construct variable. This explains that all indicators in each construct variable meet the discriminant validity requirements.

Table 5. Validity Test of AVE Square Roots

Variable	Product	Promotion	Value	Distribution	Decision
Product	0,785	0,412	0,481	0,459	0,602
Promotion	0,412	0,753	0,482	0,484	0,586
Value	0,481	0,482	0,809	0,483	0,64
Distribution	0,459	0,484	0,483	0,838	0,585
Decision	0,602	0,586	0,64	0,585	0,845

Source: Research Processed Results, 2021

In the table above it can be seen that the square root value of AVE along the diagonal line has a greater correlation between one construct and another so it can be concluded that the construct has a good level of validity.

Reliability Test

Table 6. Cronbach's Alpha Results

	Cronbach's Alpha	Description
Product	0,922	Reliable
Promotion	0,941	Reliable
Value	0,735	Reliable
Distribution	0,785	Reliable
Decision	0,864	Reliable

Source: Research Processed Results, 2021

In the table above it can be seen that all the values of Cronbach's alpha construct variables are above 0.70. This explains that all construct variables meet the reliability requirements.

Table 7. Composite Reliability Results

	Composite Reliability	Criteria	Description
	Composite Reliability	Criteria	Description
Product	0,935	> 0.70	Reliable
Promotion	0,948	> 0.70	Reliable
Value	0,85	> 0.70	Reliable
Distribution	0,875	> 0.70	Reliable
Decision	0,909	> 0.70	Reliable

Source: Research Processed Results, 2021

In the table above it can be seen that all values of the composite reliability of the construct variables of the research variables are above 0.70. This explains that all construct variables meet the reliability requirements.

Structural Model Testing (Inner Model)

Table 8. R – Square results

Model structure	R-squared Coefficients	Adjusted R-squared Coefficients
Decision	0.632	0.622

Source: Research Processed Results, 2021

In the table above, it can be obtained that the value of Adjusted R Square for the decision-making variable is 0.622. This means that 62.2% of the voting decision variable is influenced by product, promotion, price and political distribution. While the remaining 37.8% is influenced by other variables not included in this study.

Average Average path coefficient (APC)=0.247, P<0.001 Average R-squared (ARS)=0.632, P<0.001 Average adjusted R-squared (AARS)=0.622, P<0.001

block VIF (AVIF)=1.691, acceptable if <= 5, ideally <= 3.3

Average full collinearity VIF (AFVIF)=1.860, acceptable if <= 5, ideally <= 3.3

Tenenhaus GoF (GoF)=0.642, small >= 0.1, medium

>= 0.25, large >= 0.36

Sympson's paradox ratio (SPR)=1.000, acceptable if

>= 0.7, ideally = 1

R-squared contribution ratio (RSCR)=1.000, acceptable if >= 0.9, ideally = 1

Statistical suppression ratio (SSR)=1.000, acceptable if >= 0.7

Nonlinear bivariate causality direction ratio (NLBCDR)=1.000, acceptable if >= 0.7

Table 9. Value of Fit Indicators and Quality Indexes
Source: Research Processed Results, 2021

1) Average path coefficient (APC)

The average path coefficient (APC) value was obtained of 0.247 with a p-value <0.001, so it can be interpreted that the research model has good fit.

2) Average R-squared (ARS)

The average R-squared (ARS) value is 0.632 with a p-value <0.001, so it can be interpreted that the research model has a good fit.

3) Average adjusted R-squared (AARS)

An average adjusted R-squared (AARS) value of 0.622 was obtained with a p-value <0.001, this could mean that the researcher's model had good fit.

4) Average block VIF (AVIF) & Average full collinearity VIF (AFVIF)

The average variance inflation factor (AVIF) value is 1.691 and the average full collinearity variance inflation factor (AFVIF) is 1.860 <3.3, which means that there is no multicollinearity problem between indicators and between exogenous variables.

5) Tenenhaus GoF (GoF)

The tenenhaus goodness of fit value was obtained of 0.642 > 0.36 which indicates that the predictive power of the model is large or the fit model is very good.

6) Sympson's paradox ratio (SPR), R-squared contribution ratio (RSCR), Statistical suppression ratio (SSR), Nonlinear bivariate causality direction ratio (NLBCDR)

To evaluate the quality indexes, the Symson's paradox ratio (SPR) index is 1,000 > 0.70 (ideal), the R-squared contribution ratio (RSCR) is 1,000 > 0.90 (ideal), the statistical suppression ratio (SSR) is 1,000 > 0.70 (ideal) and the nonlinear bivariate causality direction ratio (NLBCDR) value is 1,000 > 0.70 which means that the indices have no causality problem in the model.

Table 10. Performance Importance Analysis (PIA)

No	Statement	Means	Loading Factor	Quadrant
No	Statement	Means	Loading Factor	Quadrant
1	Political candidate platform	3,993	0,753	II
2	Past records	4,31	0,811	I
3	Formal education	3,979	0,758	II
4	Certain age considerations	3,71	0,754	II
5	Lead experience	4,2	0,755	II
6	Good governance	4,214	0,872	I
7	Good morals	4,428	0,853	I
8	The level of ideological similarity	3,786	0,785	I
9	Political promise	3,703	0,713	II
10	Advertisement via tv	3,034	0,781	III
11	Advertising via radio	2,931	0,79	IV
12	Ads through newspapers	3,034	0,806	IV
13	Advertising through magazines	3,014	0,812	IV
14	Publication via television	3,545	0,762	III
15	Advertising via radio	3,331	0,752	III
16	Ads through newspapers	3,283	0,749	III
17	Advertising through magazines	3,248	0,749	III
18	Debate events	3,662	0,723	II
19	Celebrity	2,69	0,707	III
20	Publication via	3,283	0,746	III
20	Facebook	3,283	0,746	III
21	Publication via Twitter	3,345	0,734	III
22	Publication via	3,469	0,708	III
22	Youtube	3,469	0,708	III
23	Publication via Blog	3,352	0,718	III
24	Economy Price	4,021	0,788	I
25	Psychological price/protection value	3,614	0,873	IV
26	National image	4,4	0,763	II
27	Live meeting	3,655	0,744	II
28	Indirect interaction	3,4	0,877	IV
29	Interaction by other parties	3,614	0,886	IV
30	Attention	3,738	0,782	II
31	Interest	3,428	0,77	III
32	Intension of choosing a	4,338	0,912	I
32	candidate	4,338	0,912	I
33	The act of choosing a candidate	4,407	0,907	I
Average		3,64118	0,78464

Source: Research Processed Results, 2021

After testing the hypothesis, the following table summarizes the hypotheses that have been tested:

Table 11. Hypothesis test results

No	Influence	Path Coefficients	P	Decision
No	Influence	Path Coefficients	values	Decision
H1	Product → Decision	0,294	<0,001	Significant
H2	Promotion → Decision	0,176	0,014	Significant
H3	Value → Decision	0,308	<0,001	Significant
H4	Distribution → Decision	0,208	0,005	Significant

Source: Research Processed Results, 2021

H1: A path coefficients value of 0.294 is obtained, which means that for every 1 unit increase in perceptions of political products, the decision to vote will increase by 0.294 and vice versa assuming other variables are constant. Then a p value <0.001 is obtained, which means that political products have a significant effect on voting decisions.

H2: A path coefficient value of 0.176 is obtained, which means that every 1 unit increase in perceptions of political promotion will increase the decision to vote by 0.176 and vice versa assuming other variables are constant. Then obtained a p value of 0.014 <0.05 which means that political promotion has a significant effect on the decision to vote.

H3: A path coefficients value of 0.308 is obtained, which means that for every 1 unit increase in perceptions of political prices, then it will increase the decision to choose by 0.308 and vice versa assuming other variables remain. Then a p value <0.001 is obtained, which means that political prices have a significant effect on the decision to vote.

H4: A path coefficients value of 0.208 is obtained, which means that each increase in perception of political distribution by 1 unit will increase the decision to vote by 0.208 and vice versa assuming other variables are constant. Then a p value of 0.005 <0.05 is obtained, which means that political distribution has a significant effect on voting decisions.

CONCLUSION

Based on the result of study, the conclusion are; (1) political products have a significant effect on the decision to elect repeat voters in general elections in Indonesia, a case study on FEB students at the Universitas Riau, (2) political promotion has a significant effect on the decision to elect repeat voters in general elections in Indonesia, a case study of FEB students at the Universitas Riau, (3) political prices have a significant effect on the decision to vote for recurring voters in general elections in Indonesia, a case study on FEB students at the Universitas Riau, and (4) political distribution has a significant effect on the decision to elect repeat voters in general elections in Indonesia, a case study on FEB students at the Universitas Riau.

Moreover, this research is expected to be a reference and basis for further research. For those who want to do further research, they can add other variables that can influence the decision to choose, such as social media, promotion mix, and others so that the results obtained are even better. Because in this study, 62.2% of the voting decision variables were influenced by product, promotion, price and political distribution. While the remaining 37.8% is influenced by other variables not included in this study.

REFERENCES

Alwie, A. F. (2020). Political Marketing “Strategi Pemasaran dalam Mempengaruhi Keputusan Pemilih, Manggu Makmur Tanjung Lestari, Bandung.

Averus, A., & Alfina, D. (2020). Partisipasi Politik Dalam Pemilihan Kepala Desa. Moderat: Jurnal Ilmiah Ilmu Pemerintahan, 6(3), 585–610.

Ferdinand, A. (2014). Metode Penelitian Manajemen: Pedoman Penelitian untuk Penulisan Skripsi Tesis dan Desrtasi Ilmu Manajemen.

Filan, F., & Firdaus, S. U. (2022). Kedaulatan Rakyat Dalam Pemilihan Kepala Desa. Souvereignty, 1(1), 166–170.

Ghozali, I. (2008). Structural equation modelling, Edisi II. Semarang: Universitas Diponegoro.

Nanda, V. S. (2017). Pengaruh Kesadaran Politik Warga Masyarakat terhadap Tingkat Partisipasi Politik dalam Pemilihan Kepala Daerah (Studi Deskriptif di Kabupaten Majalengka). FKIP Unpas.

Pramono, A. (2018). Ideologi dan Politik Hukum Pancasila. Gema Keadilan, 5(1), 74–84.

Qurbani, I. D. (2012). Politik hukum pengelolaan minyak dan gas bumi di Indonesia. Arena Hukum, 5(2), 115–121.

Saputra, R. (2017). Partisipasi Politik Pemilih Pemula Pada Pemilihan Presiden Di Kecamatan Mandau Kabupaten Bengkalis Tahun 2014. Jurnal Online Mahasiswa (JOM) Bidang Ilmu Sosial Dan Ilmu Politik, 4(1), 1–12.

Setiawan, H. (2017). Analisis Pengaruh Political Marketing Mix (Bauran Pemasaran Politik Terhadap Keputusan Masyarakat Kota Pontianak Memilih Wali Kota Pontianak Periode 2013-2018. Jurnal Manajemen Fakultas Ekonomi dan Bisnis. Universitas Tanjungpura.

Shofi, S., Seroja, D. P., Fajerin, M. N., Caesar, D., Sari, J. I., Zulfikar, A., Kristal, F. B., Nurmansyah, A. R. N., Winarto, D. N., & Putri, J. (2020). Kebebasan Media Mengancam Literasi Politik (Vol. 1). Prodi Ilmu Komunikasi Universitas Muhammadiyah Malang

Silalahi, T. S. (2019). Pemuda Milenial. CV Jejak (Jejak Publisher).

Sugiono, A, (2013). Strategic Political Marketing. Ombak dua. Yogyakarta.

Ta’dung, P. R. (2017). Pengaruh Pendidikan Formal Terhadap Partisipasi Politik Masyarakat Di Kampung Kama Distrik Wamena Kabupaten Jayawijaya. Jurnal Politik Pemerintahan, 1(1), 1–6.