Increasing interest in carbon storage and fluxes in river systems opens space for comparison of methods for the determination of total organic carbon (TOC) and inorganic carbon (TIC) in soil/sediment samples. We analysed 150 soil (leptosols and fluvisols) and sediment samples from three rivers using methods mostly used in studies. Widely used dry combustion method (with CO2 detection) for determination of TOC and TIC was compared with other methods for determination TOC (modified Walkley–Black method), total organic matter (TOM determined by loss on ignition) and TIC (calcimeter, loss on ignition). Furthermore, we proposed a procedure for calculation TOC from TOM together with relative standard deviation resulting from this conversion. Our results showed that the modified Walkley–Black method provides almost identical results of TOC as dry combustion and therefore can be used as an alternative method for the determination of TOC. A strong correlation was found between TOC and TOM (R2 = 0.97 for both methods for determination TOC). The mean relative deviation of TOC values resulting from calculation TOC based on TOM varied between 10.1% and 14.5%. Thus, the calculation TOC from TOM is possible, but only with deviations referred to above. We suggest that for samples with TOM values below 1%, the TOC values cannot be reliably calculated, and in this case, the dry combustion or modified Walkley–Black method must be used for TOC determination. TIC values determined by the three methods vary to a greater extent than those of organic carbon. We assume that in the case of low TIC content (below 1%), various methods can give markedly different results. Due to the higher relative standard deviation in case of dry combustion, it is probably better to use a calcimeter for the determination of TIC in samples with TIC content below 1%.