In the deep end: pooling data and other statistical challenges of zoo and aquarium research

摘要: Zoo and aquarium research presents many logistic challenges, including extremely small sample sizes lack of independent data points, which lend themselves to the misuse statistics. Pseudoreplication pooling are two statistical problems common in biological sciences. Although prevalence these other miscues have been documented fields, little attention has paid practice statistics field zoo biology. A review articles published journal Biology between 1999–2004 showed that approximately 40% 146 utilizing inferential during span contained some evidence pseudoreplication or data. Nearly 75% studies did not provide degrees freedom for all 20% report test statistic values. level this dataset is outside levels found branches biology, it does indicate challenges dealing with appropriate analysis studies. The standardization techniques deal methodological populations can help advance by guiding production applied This study recommends issues, complete disclosure manipulation reporting values, checking control institutional effects models, avoidance pseudoreplicated observations. Additionally, biologists should seek out models such as hierarchical factorial randomization tests supplement their repertoire t-tests ANOVA. These suggestions intended stimulate conversation examination current use biology an effort develop more consistent requirements publication. Biol 0:1–14, 2006. © 2006 Wiley-Liss, Inc.