To sample or not to sample? That is the question ... for the vegetation scientist

摘要: Lajer (2007) raised the problem of using a non-random sample for statistical testing plant community data. He argued that this violates basic assumptions tests, resulting thus in non-significant results. However, huge part present-day knowledge vegetation science is still based on non-random, preferentially collected data communities. I argue that, given inherent limits preferential sampling, change approach now necessary, with adoption sampling random principles seeming obvious choice. complete transition to random-based designs limited by yet undefined nature communities and diffused opinion have discrete nature. Randomly searching such entities almost impossible, their dependence scale observation, plot size shape, need finding well-defined types. conclude only way solve conundrum consider study as operational units. If are defined operationally, they can be investigated proper techniques analyzed adequate tools.