تقدير معلمات Meta Regression مع تطبيق عملي

RP- Meta Regression Parameters Estimation with an Application.pdf

Meta-regression can be applied to a wide range of research questions and disciplines. Such as medicine, psychology, education, economics and environmental science. It aims to summarize all studies in one study and wants it to have a general effect for all studies. Therefore, this thesis aimed to estimate a meta-regression model with random effects on a group of (17) studies on the BCG vaccine for the purpose of knowing the general effect of the vaccine using seven methods. using seven methods, which are the restricted maxim likelihood method and the method Weighted least squares method (Wietghed Least square method), Baysian method, (Hedges method), (Hunter-Schmidt method), (DerSimonian-Laird method) and (Sidik-Jonkman) method, and the comparison between the methods using the standard of standard error. The thesis concluded that the Bayes method recorded the least standard error with an effect size of (0.633) with a significance level less than 0.01, followed by the Hedges method, then the Weighted Least Squares method (WLS), then the Restricted m (RMLE) maximum Likelihood method, and then the method (Hunter-Schmidt) followed by (DerSimonian-Laird) method and finally the Sidik-Jonkman method. All methods were significant and had a significant effect size at a significance level of less than 0.01. Thus, the (BCG) vaccine has an effect on recovery from pulmonary tuberculosis in children and adults by (63%). %).