The choice of the statistical test. The most suitable statistical test is described according to the type of variable, whether or not it is paired data and follow normal distribution. The most common situations will be those of comparison of means,…
Publication bias. We describe publication bias in meta-analysis and the most used methods that are empoyed for its study, both graphical and numerical.
Study of heterogeneity in meta-analysis. Numerical (Q, I, H) and graphical methods (Galbraith and L’Abbé) for the study of heterogeneity of the primary meta-analysis studies.
Simple regression. The concept of regression is defined and the most used simple regression models are detailed: linear, logistic and proportional hazards regression.
Outliers. Outliers are values that seems inconsistent with the rest of the values, according to the assumed probability model that they follow in the sample.