Special Issue-Application of Structural Equation Modeling in Business, Sport, Tourism, Recreation, Education, and Language Studies ABAC Journal Vol.43 No.4 (October-December 2023, pp - ) https://doi.org/10.59865/abacj.2023.54 BEYOND FACTORS: IGSCA-SEM’S APPLICATION IN THE CONTEXT OF CANNABIS TOURISM Chichaya Leruksa 1 , Pongphan Sathatip 2 , Supawat Meeprom 3,* Abstract Structural Equation Modeling (SEM) is a statistical technique that is used to model the relationships among hypothetical constructs investigated by researchers. SEM can be broadly classified into two main approaches: factor-based (or covariance-based) SEM and component- based (or variance-based) SEM. Factor-based SEM is particularly well-suited for analyzing constructs that resemble factors, while component-based SEM is designed for composites or components. Historically, in the field of tourism research, there has been a tendency to treat factor models as a statistical proxy for all types of hypothetical constructs. However, when the hypothetical construct is incorrectly modeled as a factor instead of a composite, which is its appropriate representation, it can result in bias in parameter estimates. The information presented in this study highlights that this practice has persisted even in top-tier tourism journals, including articles published in the ABAC journal. Contemporary practices that align with the current research landscape in tourism are synthesized. These practices acknowledge that hypothetical constructs can either be factors or components. To illustrate this, a hypothetical example related to cannabis tourism is used, modelling it using mixed constructs based on IGSCA-SEM. Researchers are consequently encouraged to employ SEM, particularly when aiming to publish in the ABAC journal, to enhance their methodological rigor by adopting the recommended practices outlined. Keyword: Integrated Generalized Structured Component Analysis (IGSCA), SDG 8, Structural Equation Modeling, Cannabis Tourism. INTRODUCTION Structural Equation Modeling (SEM) is widely popular in the field of social sciences research. Its popularity is largely attributed to its capability to simultaneously test multiple hypotheses involving latent variables. In the SEM framework, there are two primary types of variables: latent variables and observed 1 Dr. Chichaya Leruksa is currently working as a lecturer in the Department of Tourism Industry Management, Khon Kaen University, Thailand. She obtained a Ph.D. in Tourism Management from Khon Kaen University, Thailand. 2 Dr. Pongphan Sathatip is currently working as a lecturer in Hospitality and Event Management Department, Faculty of Business Administration and Accountancy, Khon Kaen University, Thailand. He obtained a Doctoral degree in Hotel and Tourism Management from The Hong Kong Polytechnic University, Hong Kong. 3,* Assoc Prof. Dr. Supawat Meeprom is currently working as a lecturer in the Hospitality and Event Management Department, Faculty of Business Administration and Accountancy, Khon Kaen University, Thailand. He obtained a Ph.D. in Marketing and Event Management from Macquarie University, Sydney, Australia. Email Supame@kku.ac.th variables. Observed variables are those directly measured by researchers through constructed survey instruments, while latent variables are those that cannot be directly measured but are assessed through a set of questions or indicators. For instance, academic English proficiency can be assessed through a set of shared indicators, such as TOEFL scores, IELTS scores, and TOEIC