Underappreciated problems of low replication in ecological field studies

文献类型： 外文期刊

第一作者： Lemoine, Nathan P.

作者： Lemoine, Nathan P.;Hoffman, Ava;Felton, Andrew J.;Baur, Lauren;Chaves, Francis;Gray, Jesse;Yu, Qiang;Smith, Melinda D.;Yu, Qiang

作者机构：

关键词： Bayesian statistics;LASSO regression;power;priors;ridge regression;Type M error;Type S error

期刊名称：ECOLOGY （ 影响因子：5.499； 五年影响因子：6.001 ）

ISSN：

年卷期：

页码：

收录情况： SCI

摘要： The cost and difficulty of manipulative field studies makes low statistical power a pervasive issue throughout most ecological subdisciplines. Ecologists are already aware that small sample sizes increase the probability of committing Type II errors. In this article, we address a relatively unknown problem with low power: underpowered studies must overestimate small effect sizes in order to achieve statistical significance. First, we describe how low replication coupled with weak effect sizes leads to Type M errors, or exaggerated effect sizes. We then conduct a meta-analysis to determine the average statistical power and Type M error rate for manipulative field experiments that address important questions related to global change; global warming, biodiversity loss, and drought. Finally, we provide recommendations for avoiding Type M errors and constraining estimates of effect size from underpowered studies.

分类号： X17

  • 相关文献

[1]Improving accuracy of genomic prediction by genetic architecture based priors in a Bayesian model. Gao, Ning,Li, Jiaqi,He, Jinlong,Xiao, Guang,Luo, Yuanyu,Zhang, Hao,Chen, Zanmou,Zhang, Zhe,Gao, Ning,Zhang, Zhe. 2015

作者其他论文 更多>>

微信小程序