Autor: |
Anekawati, Anik, Hidayat, Syaifurrahman, Rofik, Mohammad |
Předmět: |
|
Zdroj: |
AIP Conference Proceedings; 2024, Vol. 3095 Issue 1, p1-8, 8p |
Abstrakt: |
When the sample unit is a location that possibly has a spatial effect and at the same time involves latent variables, then spatial structural equation modeling (SEM) is more appropriate. The spread of COVID-19 is certainly influenced by the affected location. If one of the sub-districts has a family infected by COVID-19, then the family in the adjacent sub-district is the most highly likely to be infected as well. Questionnaire survey data were 7,243 heads of families in 27 sub-districts in Sumenep District. The first step is to evaluate the model on SEM, both outer and inner models. Spatial modeling in this study employs queen contiguity weights. Spatial SEM modeling is run by modeling the variables in the spatial model, but the dependent and independent variables are replaced by factor scores. The factor score is the estimation result of latent variables using the SEM-PLS method. The dependency test on factor scores was carried out to obtain a spatial model of the inner model on SEM using the Lagrange multiplier test. The results of the spatial dependence test on the family resilience model against the potential spread of COVID-19 led to an autoregressive spatial model of error (SERM-SEM) at a significance level of = 5%. The variables of family members' commitment to Covid-19 prevention and the quality of the home environment affect the zone status, with the negative coefficient variables of family members' commitment. R square value is 93.54%, and AIC is 12.77. [ABSTRACT FROM AUTHOR] |
Databáze: |
Complementary Index |
Externí odkaz: |
|