Comparison of Cluster Validity Index and Distance Measures Using Integrated Cluster Analysis and Structural Equation Modeling the Warp-PLS Approach

Abstract

This study wants to compare the Integrated Cluster Analysis and SEM model of the Warp-PLS approach with various cluster validity indices and distance measures on Service Quality, Environment, Fashions, Willingness to Pay, and Compliant Paying Behavior of Bank X Customers. The data used in this study are primary. The variables used in this study are service quality, environment, fashion, willingness to pay, and compliance with paying behavior at Bank X. The data were obtained through a questionnaire with a Likert scale — measurement of variables in primary data using the average score of each item. The sampling technique used was purposive sampling. The object of observation is the customer as many as 100 respondents. Data analysis was carried out quantitatively, and a descriptive analysis was carried out first. An Integrated Cluster Analysis and SEM analysis of the Warp-PLS approach was carried out with the average linkage method on various cluster validity indices and three distance measures. The Warp-PLS approach's integrated cluster and SEM model with the Gap Index, Index C, Global Sillhouette, and Goodman-Kruskal with the Manhattan Distance are better than the Gap, Index C, Global Sillhouette, and Goodman-Kruskal with the Euclidean and Minkowski Distance. The novelty in this research is the application of Integrated Cluster Analysis and SEM of the Warp-PLS approach to compare 4 cluster validity indices, namely Gap Index, C Index, Global Sillhouette, and Goodman-Kruskal, and three distance measures, namely Euclidean, Manhattan, and Minkowski distances simultaneously.