Used in meta-analysis to display the relationship between effect size and study precision
In an attempt to detect what was originally called publication bias but is now more usually called small study effects a plot was often used to display the primary studies. Conventionally the study effect size is plotted along the $x$-axis and some function of study precision on the $y$-axis. If there is no relationship the plot resembles a funnel with the wide part down. The plot may be supplemented by regression tests. Caution is needed with the plot if either there are few studies or there is much heterogeneity or both.
