The thermal conductivity probe is widely used to determine the thermal conductivity of foods; however, its accuracy is limited by a number of factors. In this paper the influence of particle size (relative to probe diameter) on measurement accuracy was examined by measuring the thermal conductivity of sucrose of varying grades, and coffee (both whole beans and ground) using two different methods; the thermal conductivity probe and a transient comparative method. Results showed that the measurement methods were in agreement for the small particulates; however, for the larger particulates the probe method appeared to under-predict thermal conductivity. A possible explanation for this observation is that the probe causes a local variation in the porosity of the sample due to the alteration of the packing density. A suggested rule-of-thumb is that thermal conductivity probes may not be suitable for particulate materials where the particle diameter is greater than the probe diameter. (C) 2014 Elsevier Ltd. All rights reserved.