Questions Once effects of hydrological and chemical soil properties have been accounted for, does soil microbial diversity contribute to explain change in plant community structure (i.e. within-site β-diversity)? If so, at what spatial scale does microbial diversity operate? Location La Mina in Moscosa Farm, Salamanca, western Spain (dehesa community) and Laguna Larga in the Urbión Peaks, Soria, central-northern Spain (alpine grassland). Methods The abundance of vascular plant species, soil Gram-negative microbial functional types and soil chemical properties (pH, available P, extractable cations) were sampled at both sites, for which hydrological models were available. RDA was used to partition variation in plant community structure into hydrological, chemical and microbial components. Spatial filters, arranged in scalograms, were used to test for the spatial scales at which plant community structure changes. Results In the case of the dehesa the diversity of soil Gram-negative microbes, weakly driven by soil pH, contributed to a small extent (adj-R2 = 2%) and at a relative medium spatial scale to explain change in plant community structure. The abundance of a few dehesa species, both annual (Trifolium dubium, Vulpia bromoides) and perennial (Poa bulbosa, Festuca ampla), was associated with either increasing or decreasing soil microbial diversity. In the alpine meadow the contribution was negligible. Conclusions Microbial diversity can drive community structure, although in the hierarchy of environmental factors structuring communities it appears to rank lower than other soil factors. Still, microbial diversity appears to promote or restrain individual plant species. This paper aims to encourage future studies to use more comprehensive and insightful techniques to assess microbial diversity and to combine this with statistical approaches such as that used here.