MODEL INTEGRATIF CITRA KANDIDAT DAN KAMPANYE POLITIK TERHADAP MINAT DAN PERILAKU MEMILIH PADA PEMILIHAN GUBERNUR MALUKU = AN INTEGRATIVE MODEL OF CANDIDATE IMAGE, POLITICAL CAMPAIGNS, CULTURAL VALUES, AND PSYCHOLOGICAL DETERMINANTS OF VOTING INTENTION IN THE MALUKU GUBERNATORIAL ELECTION

Political marketing has become a crucial strategic approach in democratic contests, especially since the 1998 Reformation opened up wider political participation in Indonesia. The 2024 Maluku Gubernatorial Election in Ambon City displayed intriguing dynamics with three pairs of candidates from different backgrounds, notably where the pair Hendrik Lewerisa-Abdullah Vanath (number 3) successfully defeated the incumbent despite having a lower level of recognition. This phenomenon indicates complex factors influencing voter behavior, including candidate image, political campaign strategies, and strong local cultural values such as pela-gandong prevalent in Maluku society. This study aims to test an integrative model combining candidate image, political campaign, and cultural values on attitude, interest, and voting behavior in the Maluku Gubernatorial Election. Although the Theory of Planned Behavior (TPB) has been widely used, its application in local political contexts often fails to fully explain phenomena influenced by cultural and social factors. This research seeks to bridge this gap by adapting a combination of the marketing mix theory and TPB within the context of a regional head election (Pilkada). The study was conducted through a two-stage longitudinal study: Study I (pre-election) tested the influence of external variables (candidate image, political campaign, and cultural values) on TPB variables (attitude, subjective norms, and perceived behavioral control) and voting interest; and Study II (post-election) examined the influence of voting interest on actual voting behavior. The research population was the community of Ambon City, spread across five districts. Non-probability sampling using a purposive sampling method was employed, targeting voters who had Pela-Gandong kinship ties with one of the candidates. Data analysis used two methods: Partial Least Square-Structural Equation Modeling (PLS-SEM) for Study I and Binary Logistic Regression for Study II. The results statistically supported all hypotheses. Significantly, the political campaign had the strongest influence on voter attitude (coefficient 0.406, t=11.176, p<0.001), while cultural values showed the strongest influence on subjective norms (coefficient 0.627, t=24.032, p<0.001). These findings confirm that voting behavior in Maluku is not solely driven by personal factors like attitude but is also shaped by profound social and cultural influences. Furthermore, voting intention was proven to be a strong predictor of actual voting behavior. This research has both practical and theoretical implications. Practically, the findings offer insights for campaign teams to design more effective strategies, focusing on building a positive image, transparent campaigning, and integrating local cultural values. Theoretically, the study provides strong empirical evidence that voter behavior models must account for the unique cultural and social factors of a region, thereby enriching the understanding of TPB in a non-Western context.