Freshwater fishes are among the most endangered groups in North America. Hydrologic alteration, land use changes, and invasions have been identified as probable contributors to freshwater biodiversity losses. Because these potential causes of decline are frequently correlated, it is difficult to separate direct from indirect effects of each factor and to appropriately rank their importance for conservation action. In particular, it has recently been proposed that invasive species may be merely “passengers” (of habitat change) rather than “drivers” of biodiversity loss in most ecosystems. I used path analysis to assess the relative fit of the “driver” vs. “passenger” models of invasions as contributors to biodiversity loss of freshwater fishes in three regions: California, USA; Pennsylvania, USA; and North America as a whole. The results for California indicated that nonindigenous fish diversity was the primary direct driver of biodiversity loss among native fishes, while the effects of hydrologic and land use changes were largely indirect, through their positive association with invasions. The same pattern held for the North American data as a whole, although more weakly, possibly reflecting high variability among regions. Preliminary results for Pennsylvania supported the passenger model, suggesting that habitat change is a more significant driver than invasions of fish diversity losses in eastern North America. These results suggest, first, that invasive fishes probably do drive fish diversity loss in some (perhaps largely western) regions; and second, that careful, regional analyses of potential contributors to biodiversity losses are needed to guide conservation action.