Composite Model of Marketing Mix and Brand Equity for Purchasing Motorcycles

Abstract

The current study investigates the effect of a composite model of the marketing mix and brand equity on motorcycle purchase decisions. For second-order, the marketing mix and brand equity were constructed using a disjointed two-state approach and a formative-formative weighting scheme. The purchasing decision was measured as a latent variable using a consistent partial least square (PLSc) model. The data was collected from 148 people who have bought motorcycles in Thailand. The results show that the lower-order constructs demonstrate how well they are organized by the indicators, as determined by the model fit indexes. To begin, the hypothesis regarding the three model fitting indexes is rejected. Thus, all bias indicators with shallow values and a negative symbol were removed from the composite model. Furthermore, multicollinearities exist between PLACE indicators, and they are addressed by removing the indicator with the highest variance inflation factors (VIF). Concerning purchasing decisions, the PLSc model with an indicator loading of less than 0.708 was eliminated. As a result, all three total model fit hypotheses returned to be accepted. For the higher-order construct, the composite model created a marketing mix positively linked to brand equity and purchasing decisions, and brand equity positively correlates with purchasing decisions. The hybrid model can generate model fit indexes for both the first and second construct.