Estimating herbivore density is an important part of understanding their impact on vegetation.  Many studies have been carried out on the impact of reindeer and other herbivores on arctic and sub-arctic vegetation, but they are difficult to compare as they typically use different methods to estimate herbivore activities.  The aim of this study was to compare three methods that were based on the recent International Tundra Experiment herbivory protocol to measure the activities of three herbivore groups: reindeer, rodents and invertebrates. The robustness of the methods themselves was then evaluated.  Fieldwork was carried out at 12 sites in the Fennoscandian mountain area, with controls inside reindeer exclosures.  The results showed that the methods were the most robust when measuring reindeer activities.  The reindeer measurements were also well correlated with a reindeer-density estimate calculated from official reindeer population data.  This study recommends considering the use of photographs to increase the time-efficiency of pellet-counts.  The rodent activity estimates were good, but the patterns inside exclosures differed to the patterns outside exclosures.  The results for invertebrates were deemed to be less reliable as the measurements for one method were not recorded at an appropriate scale.  In conclusion, the findings of this study will help improve the comparability of future studies on the impact of reindeer herbivory and other herbivores, and gives suggestions for more accurate ways of measuring herbivore pressure in arctic and sub-arctic vegetation.