The effects of consumer confusion on hotel brand loyalty: an application of linguistic nonlinear regression model in the hospitality sector
Kurtulmusoglu, Feride Bahar
Atalay, Kumru Didem
MetadataShow full item record
The aim of the study is to estimate the interaction and quadratic relationships between dimensions by estimating a model for the confusion dimensions that affect hotel brand loyalty, thus providing the interested parties with a perspective and direction regarding consumer confusion. This study also aimed to strengthen the use of FLS in the field of social sciences and will use this method to transform the discrete ordinal variable into a continuous variable while preserving the semantic meaning. Four hundred and six individuals participated in the study. Hypotheses demonstrating the interaction and quadratic effects between the continuous variables have been analysed using nonlinear multiple regression analysis. This study proposes a survey-based method to estimate a model for the confusion dimensions that affect hotel brand loyalty. The results demonstrated that ambiguity confusion, overload confusion, similarity confusion, quadratic effect of similarity confusion and interaction of ambiguity, overload and similarity confusion decrease the hotel brand loyalty. Also, quadratic effect of ambiguity confusion, interaction of ambiguity and overload confusion, interaction of overload and similarity confusion, interaction of ambiguity and similarity confusion increase the hotel brand loyalty. Despite its importance for marketing and consumer behaviour, the definition, measurement, dimensions and existing results of consumer confusion have begun to be discussed and examined recently in a limited scope. Studies have demonstrated that consumer confusion about tourism products is a non-functional and under-evaluated area but also is utmost prominent for tourism product. This study aimed to obtain a stronger model in which all the interactions between variables and their (quadratic) increasing effects are considered using a nonlinear regression model.