Structural equation models and an application to the Turkish commodity exchange data

In this study, the satisfaction levels of investors operating in Batman and Siirt provinces towards the Turkish Commodity Exchange (TÜRİB) and licensed warehousing system were analyzed using structural equation modeling. The main objective of the study is to reveal the determinants of satisfaction in line with investors' expectations, experiences and trust levels in the system. In the study, survey data obtained through face-to-face interviews with a total of 388 investors in Batman and Siirt were used. Exploratory factor analysis (EFA) and confirmatory factor analysis (CFA) were applied in the data analysis process. As a result of EFA, five factors (effectiveness of licensed warehousing, warehousing ınfrastructure and user competence, supporting ınfrastructure and policies, ease of product supply, reliability and access) were identified; 63,887% of the total variance was explained. The reliability of the questionnaire scale was tested with Cronbach's Alpha value and was found to be 0.795 and was considered reliable. As a result of CFA, the goodness of fit indices of the model (RMSEA=0,077, CFI=0,944, TLI=0,919) were found to be acceptable. Three of the five hypotheses tested within the scope of the structural equation model were found to be statistically significant. While the factors of effectiveness of licensed warehousing and supportive infrastructure and policies positively affect service quality and satisfaction (SQS), the factor of member orientation and institutional representation has a negative effect on SQS. On the other hand, the effects of ease of product supply and reliability and access factors on SQS were not statistically significant. The explanatory level of the model was determined as R²=0,769. The research findings aim to provide guiding recommendations to policy makers and practitioners for the development of TURIB and licensed warehousing system.