Figure 1.

Overview of literature search results and categorization of studies. After literature searching and screening, the 102 studies that met our inclusion criteria were divided into experimental studies, where researchers controlled the genetic diversity and parasite exposure of experimental host populations, and observational studies, where researchers measured parasitism and genetic diversity across unmanipulated populations. Experimental studies were classified by host to distinguish noncrop from crop systems. Crop studies were further divided according to whether they provided an estimate of sampling variance (“sd” for standard deviation). We outline the predictions and effect sizes for these study types. We used the log response ratio (lnRR) and Fisher'szfor statistical analyses (Table S2), but here and in the Results we present these effect sizes as the response ratio and correlation coefficient (r) because they are easier to interpret.