Assuming that agglomeration effects do matter in organic farming we analyse (a) the difficulties due to data aggregation arising when trying to statistically verify neighbourhood effects and (b) whether results can be confirmed at different spatial resolutions. Explaining the spatial distribution of organic farming in southern Germany (2007) we compare results of spatial lag models at two measurement scales. The results suggest that essential factors determining the decision to convert from conventional to organic farming are found at different spatial resolutions. The results at the lower spatial resolution are not artificially generated through the aggregation process in this case, strengthening the relevance of previous studies.
