MODEL SEM-PLS TERBAIK UNTUK EVALUASI PEMBELAJARAN MATEMATIKA DISKRIT DENGAN LMS
Research on the use of Structural Equation Modelling-Partial Least Square (PLS-SEM) related to Learning Management System LMS has developed very rapidly. However, in these studies, it was not explained how to choose the best model used to evaluate the relationship among latent variables in the model. This study aims to select the best SEM-PLS model related to evaluating the use of LMS in Discrete Mathematics learning based on the criteria of Q2, AIC, AICu, AICc, BIC, HQ, and HQc. Data obtained from a survey of 109 3rd semester students who took Discrete Mathematics courses at STMIK IKMI Cirebon using 5 latent variables. The Main Model is formed based on all research latent variables and evaluated by stages 1) PLS-Algorithm, 2) Bootstrapping and 3) Blindfolding. Based on the Main Model, 16 alternative models are created with the same manifest variables as the Main Model. The best model is determined based on the highest Q2 value, and the least AIC, AICu, AICc, BIC, HQ and HQc values. The results of the study show that the Main Model is better based on the Q2 value compared to other models in this study. Different results are obtained if the AIC, AICu, AICc, BIC, HQ and HQc criteria are used, where Model C2 and B2 are the best models based on these criteria
Copyright (c) 2019 Novi Mardiana, Ahmad Faqih
This work is licensed under a Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International License.