生态学实验“可重复”困难的原因及对策

 生态学实验“可重复”存在诸多困难。在本体论上,主要有自然的变异性以及大尺度的限制等原因,对此,采取的对策是：或者使用易于处理的生物或生态系统来阐明相关过程,或者选择那些同质性的或平衡的系统进行研究,或者模拟自然进行微宇宙实验。不过,所有这些对自然的简单化方面的选择或者对自然的简化模拟,目的是获得“可重复性”高的实验,虽然一时有效,但并非总是可行。在认识论上,生态学实验对象的复杂性、有机整体性、历史性决定了对它们的相关认识的正确性受到限制,这直接影响到实验的“可重复”,为此,准确确定实验场所、清楚界定相关概念等就成为必需,由此能够达到生态学实验的正确性（实在性）与“可重复性”的双赢。在方法论上,不完整的实验报告以及缺乏相关的方法细节,是造成生态学实验“可重复”困难的重要原因,鉴此,完善实验报告和评审体制、提供实验细节原始记录、执行严格的论文评审标准就成为必需。在价值论上,学术不端行为（如P值篡改、择优选择、结果已知之后假设等）成为实验“可重复”困难的一个重要方面,必须杜绝。

The “repeatability” difficulty in ecological experiment

The "repeatability" difficulty in the ecological experiment is an important issue. From the point of view of the ontology, mainly due to the natural variability, and partly due to the large scale, it is easy to use an organism system to illustrate the process, or to choose the homogeneity and balancing system to simulate the natural micro universe experiments. However, all these simplifications of the natural selection, or simplified simulations of the nature, although valid for a moment, are not always feasible. From the point of view of the epistemology, the ecology of the complexity of the subjects, the organic wholeness and historic nature make the validity of the understanding restricted, which directly affects the experimental "repeatability". Therefore, it is necessary to accurately determine the experiment site, and to clearly define the relevant concepts, to achieve the correctness(reality) of the ecology experiment and create a "repeatable" win-win situation. Methodologically, the incomplete experimental reports and the lack of relevant method-specific details are important causes of the "repeatability" difficulty of ecological experiments, thus it is necessary to improve the experimental report and the review system, to provide the original records of experimental details, to apply strict evaluation criteria. From the point of view of the axiology, academic misconducts, such as the P-hacking, the cherry-picking, and the hypothesizing after results(HARKing), etc., become important aspects of the experiment's "repeatability" difficulty and must be eradicated completely.