Total microbial activity (TMA) in soils is vital for understanding the roles of microorganisms in ecosystem processes. It can be defined as the sum of physiological activities of all the microbes at a given moment. As TMA is difficult to measure directly, a series of proxies, such as respiration rates, growth rates, and cellular RNA concentration, have been proposed. Here, methods used to measure soil TMA are synthesized and compared. (1) Respiration may be the process most closely related to life activities. Thus, respiration rates are the most commonly used proxies of soil TMA. The main limitation is that current methods to determine respiration rates usually cannot accurately reflect actual respiration rates. When respiration rates are measured using CO2 production or O2 consumption rates, they indicate carbon mineralization or aerobic respiration rates, respectively. (2) Microbes with higher growth rates are usually more active. Thus, growth rates are also widely used to indicate soil TMA. As biomacromolecule synthesis is approximately proportional to microbial growth rates, incorporation of radioactive isotope labeled precursors (i.e., thymidine, leucine, and acetate) can be employed to estimate microbe growth rates. Generally, trace radioactively labeled precursors are added to slurries (traditional methods) or extracted microbial suspensions (Bååth's methods). After a brief incubation, microbes are killed and the corresponding biomacromolecules are extracted to measure their radioactivity. Thymidine and leucine incorporation are commonly used to measure heterotrophic bacterial growth rates, while acetate incorporation is used to estimate growth rates of saprotrophic fungi. Radioactive isotope labeling methods are robust tools to estimate the growth rates of soil microbes. However, one critical problem is that TMA includes both growth activity and non-growth activity, whereas these methods only reflect the former. (3) Evidently, none of the methods based on respiration rates or incorporation of radioactive isotope labeled precursors can accurately link microbial activity with their identities. However, this issue can be resolved through using methods based on RNA. RNA correlates closely with protein synthesis, which is involved in most metabolic processes. Therefore, RNA concentration is assumed an ideal indicator of microbial activity. Generally, mRNA can be used to indicate the activity of specific metabolic processes, including nitrogen fixation and denitrification, whereas rRNA is a proxy of soil TMA. As cellular concentrations of small subunit rRNA (SSU rRNA) are proportional to total cellular rRNA concentrations, SSU rRNA copies can serve as an indicator of soil TMA and the ratio of SSU rRNA copies to SSU rRNA gene copies can be used to determine average microbial activity in soil. Additionally, active microbial community structure can be illustrated using profiling methods, such as clone library, T-RFLP, and high-throughput sequencing, based on SSU rRNA. These approaches can simultaneously identify soil microbes and their activity via in situ measurements. However, there is still no adequate evidence to support the assertion that the methods based on SSU rRNA can accurately reflect microbial activity, especially for non-growth activity. In conclusion, none of these methods are perfect to determine soil TMA; and a combination of suitable methods should be selected for individual ecosystems.