Plant invasion science has made a substantial progress in documenting the impacts of aliens, but comparisons with the impacts of native dominants are still rare. Further, the impacts on larger spatial scales remain poorly understood. We recorded the impacts of 10 native and nine invasive dominant plants in the Czech Republic on species richness and Shannon diversity by comparing communities with high vs. low cover of the dominant species. To estimate the impacts at the (i) population level and (ii) between-population level, we compared the Jaccard dissimilarity, nestedness and turnover of high- and low-dominance plots. Further, we calculated the Jaccard dissimilarity, nestedness and turnover between the high- and low-dominance plots within each population to express the impacts on species composition. We tested whether (i) native and invasive dominants affect the population- and between population levels of diversity by making the vegetation more homogenous; (ii) whether these effects differ between the native and alien dominants; and (iii) whether the impacts at different spatial levels are related. At the population level, high-dominance plots (with both native and alien dominants) showed higher nestedness and lower turnover compared to the low-dominance plots. Further, all plots with native dominants, both with high- and low dominance, showed higher similarity but lower nestedness than plots with alien dominants. Most importantly, high-dominance plots with native dominants were more similar to each other but showed marginally significantly lower nestedness compared to high-dominance plots with alien dominants. At the between-population level, high-dominance plots with native dominants showed a marginally significantly lower turnover compared to high-dominance plots with alien dominants. The differences in Jaccard dissimilarity, nestedness and turnover between the low- and high-dominance plots at the population level showed strong positive relations to low- and high-dominance differences at the between-populations level. Further, compositional impacts, expressed as the dissimilarity between high- vs. low-dominance plots, positively related to the plot-level impacts on Shannon diversity. Our results show that (i) both native and invasive dominants tend to reduce the diversity over larger areas and that the effect of native dominants may be even stronger, and (ii) the effects on plot-level richness and diversity cannot be easily extrapolated to larger scales but the impacts at the population- and between-populations levels are positively related.