In the face of unprecedented loss of biodiversity, cross-taxon correlates have been proposed as a means of obtaining quantitative estimates of biodiversity for identifying habitats of important conservation value. Habitat type, animal trophic level, and the spatial extent of studies would be expected to influence the strength of such correlations. We investigated these effects by carrying out a meta-analysis of 320 case studies of correlations between plant and animal species richnesses. The diversity of arthropods, herps, birds, and mammals significantly increased with plant diversity regardless of species habitat. However, correlations were stronger when plant and animal species richnesses were compared between habitats (gamma diversity) than within single habitats (alpha diversity). For arthropods, both the coefficient of correlation and the slope of the regression line were also greater for primary than for secondary consumers. These findings substantiate the use of plant species richness as an indicator of the diversity of animal taxa over space.