Background: Meta-analyses combine the results of multiple studies of a common question. Approaches based on effect size estimates from each study are generally regarded as the most informative. However, these methods can only be used if comparable effect sizes can be computed from each study, and this may not be the case because of variation in how the studies were done or because of limitations in how their results were reported. Other methods must then be used to summarise the results of these studies. One possibility is a simple vote counting method, where studies are divided by statistical significance and direction to give an overall indication of the number of studies showing an association. Preferable to vote counting, meta-analysis of P values can be undertaken using Fisher’s or Stouffer’s method. These methods have important limitations however, due to the well-known pitfalls of P values and in particular their dependence on sample size: without sample size, a given P value could have any magnitude of effect.

Development: We propose a novel plot that requires only a P value, a total sample size and a direction of effect from each study. Notably, the plot allows an approximate examination of underlying effect sizes and the potential to identify sources of heterogeneity across studies. This is achieved by drawing contours showing the range of effect sizes that might lead to each P value for given sample sizes, under simple study designs. These contours enhance the interpretability of the albatross plots, so named because the contours resemble the eponymous bird’s wings. Examples of albatross plots using real data are presented, and their production and utility are discussed.