METAPROP_ONE: Stata module to perform fixed and random effects meta-analysis of proportions
Victoria Nyawira Nyaga, Marc Arbyn, Marc Aerts · RePEc: Research Papers in Economics · 2014
This routine provides procedures for pooling proportions in a meta-analysis of multiple studies study and/or displays the results in a forest plot. The pooled estimate is obtained as a weighted average, by fitting the Logistic regression model without covariate but with an intercept, or by fitting the logistic-Normal random-effects model without covariates but with random interceptsThe confidence intervals are based on score (Wilson: Newcombe, R. G. 1998) or exact binomial (Clopper-Pearson) (Newcombe, R. G. 1998) procedures. A test for heterogeneity, i.e., whether the proportion in all studies is the same, as well as test of whether the summary proportion is equal to zero is given. Heterogeneity is also quantified using the I-squared measure (Higgins et al. 2003). Metaprop allows also for sub-group meta-analysis and produces a p-value for differences in proportions by one covariate.