Using a two-level structural equation model to study the determinants of reproductive behaviour in Giza Governorate

摘要

The purpose of this paper was to measure the quality of concordance of one- and two-level structural equation models by testing the two following statistical hypotheses: first, the structure of the two- level model is more efficient than that of the one-level model; second, economic and social determinants and intermediate variables have a significant effect on current and future fertility in Giza Governorate. A two-tier bicameral model (women, geographical area) was built on 1500 families divided equally into three districts (urban, rural, and slum). The districts were divided into 294 convergent communities according to the standard of living. M-Plus 8.0 was used to estimate the numbers of four models. The first estimated structural relationships between the determinants of fertility within the layers; the second estimated relationships between the second and third strata; the third estimated structural equations at the second level; and the fourth included the constants and a random slope in the model. The results show that the estimated parameters of the models corresponded well to the model and the multi-level model, whose indicators are proposed within acceptable limits. The final proposed model and the significance of its coefficients were tested to ensure the quality of the compatibility of the final model. The results show that the overall quality of the proposed compatibility is very high and that the indicators are within the accepted limits. The compatibility quality index is 0.99, the correlation quality is corrected by 0.98, and the Cl index of significance is 0.00. Therefore, we can say that the proposed multilevel structural model (based on the indicators of total compatibility quality) is more efficient than the one-level model, supporting the first hypothesis. The second hypothesis is that economic and social determinants and intermediate variables have a significant impact on current and future fertility. (C) 2019 Production and hosting by Elsevier B.V. on behalf of Faculty of Computers and Information, Cairo University.