Methanogenic community structure and population dynamics were investigated in two anaerobic reactors treating a dairy wastewater, an Inverted Fluidized Bed (IFB) and Expanded Granular Sludge Bed (EGSB). A combination of real-time PCR, denaturing gradient gel electrophoresis and statistical techniques was employed. Distinct methanogenic communities developed in the IFB and EGSB reactors reflecting step-wise reductions in the applied hydraulic retention time from 72 to 12 h during the 200-day trial. The aceticlastic family Methanosarcinaceae was only detected in the IFB and the order Methanomicrobiales was also much more abundant in this reactor, while the aceticlastic family Methanosaetaceae was more abundant in the EGSB. The hydrogenotrophic order, Methanobacteriales, predominated in both reactors under all applied operational conditions. Non-metric multidimensional scaling (NMS) and moving-window analyses, based on absolute and relative abundance quantification data, demonstrated that the methanogenic communities developed in a different manner in the IFB, compared to the EGSB reactor. In our study, relative abundance-based quantification by NMS and moving-window analysis appeared to be a valuable molecular approach that was more applicable to reflect the changes in the anaerobic digestion process than approaches based either on qualitative analysis, or solely on absolute quantification of the various methanogenic groups. The overall results and findings provided a comparative, quantitative and qualitative insight into anaerobic digestion processes, which could be helpful for better future reactor design and process control.