Release of PCDD and PCDF from biomass combustion such as forest and agricultural crop fires has been nominated as an important source for these chemicals despite minimal characterisation. Available emission factors that have been experimentally determined in laboratory and field experiments vary by several orders of magnitude from <0.5 μg TEQ (t fuel consumed)−1 to >100 μg TEQ (t fuel consumed)−1. The aim of this study was to evaluate the effect of experimental methods on the emission factor. A portable field sampler was used to measure PCDD/PCDF emissions from forest fires and the same fuel when burnt over a brick hearth to eliminate potential soil effects. A laboratory burn facility was used to sample emissions from the same fuels. There was very good agreement in emission factors to air (EFAir) for forest fuel (Duke Forest, NC) of 0.52 (range: 0.40–0.79), 0.59 (range: 0.18–1.2) and 0.75 (range: 0.27–1.2) μg TEQWHO2005 (t fuel consumed)−1 for the in-field, over a brick hearth, and burn facility experiments, respectively. Similarly, experiments with sugarcane showed very good agreement with EFAir of 1.1 (range: 0.40–2.2), 1.5 (range: 0.84–2.2) and 1.7 (range: 0.34–4.4) μg TEQ (t fuel consumed)−1 for in-field, over a brick hearth, open field and burn facility experiments respectively. Field sampling and laboratory simulations were in good agreement, and no significant changes in emissions of PCDD/PCDF could be attributed to fuel storage and transport to laboratory test facilities.