The current study describes quantitatively the interactions between (a) activity of extracellular enzymes: β‐glucosidase (BG), protease (Prot), alkaline phosphatase (Alk P), and acid phosphatase (Acid P); (b) intracellular activities: respiration, dehydrogenase activity (DHA); and (c) C and N contents of soil: Corg—organic carbon, Cmic—microbial biomass, ratio Cmic:Corg, Ntot—total nitrogen, and ratio C:N in earthworm‐affected and surrounding loess soil under a pear orchard. The soil tested was collected from: burrow walls (0–3 mm) and a transition zone (3–7 mm) both forming drilosphere, bulk soil (20 mm from the burrow wall), and cast aggregates. The enzyme activities were presented as total ‘tot’ (per gram of dry soil mass) and specific ‘spec’ (per Cmic). It was shown that soil Cmic, basal respiration, BGtot, Prottot, and Alk Ptot were significantly positively correlated with Corg, Cmic:Corg, Ntot, and C:N (r = 0.627–0.976). However, the specific BGspec, Protspec, and Acid Pspec were negatively correlated, while specific respiration (qCO2) and Alk Pspec were not correlated with most soil characteristics. The relationships between DHAtot, DHAspec, and all soil characteristics were positive and had the highest correlation coefficients (r = 0.703–0.989). Response ratio of the enzyme activity was more strongly affected by enzyme activity location (extracellular vs. intracellular) and presentation method (total vs. specific), than by earthworm‐affected soil compartments. The results suggest a need to consider the method of enzyme activity presentation (specific vs. total) that supports understanding ecologically relevant microbial processes in earthworm‐inhabited orchard soils.