A guide to sample size for animal-based studies / Penny S. Reynolds.
- 作者： Reynolds, Penny S.
- 出版： Hoboken, NJ : Wiley-Blackwell 2024.
- 主題： Animal experimentation--Methods. , Research--Methodology. , Animal Experimentation. , Sample Size.
- ISBN： 978-1-119-79997-9 (pbk.) :: USD71.95
- 書目註：Includes bibliographical references and index.
-
讀者標籤：
- 系統號： 005292406 | 機讀編目格式
館藏資訊
摘要註
"How large a sample size do I need for my study"? Although one of the most commonly-asked questions in statistics, the importance of proper sample size estimation seems to be overlooked by many preclinical researchers. Over the past two decades, numerous reviews of the published literature indicate many studies are too small to answer the research question and results are too unreliable to be trusted. Few published studies present adequate justification of their chosen sample sizes, or even report the total number of animals used. On the other hand, it is not unusual for protocols (usually those involving mouse models) to request preposterous numbers of animals, sometimes in the tens or even hundreds of thousands, "because this is an exploratory study, so it is unknown how many animals we will require"--
內容註
Introduction : The sample size problem in animal-based research -- Sample size basics -- Ten strategies to increase information (and reduce animal numbers) -- Introduction to pilot studies -- Operational pilots -- Empirical and translational pilots -- Feasibility calculations : Arithmetic estimation -- Feasibility calculations : Probability-based estimation -- Introduction to descriptive studies -- Confidence intervals -- Prediction intervals -- Tolerance intervals -- Reference intervals -- Hypothesis testing, power, non-centrality -- A bestiary of effect sizes -- Comparing two groups : Continuous outcomes -- Comparing two groups : Binary and count outcomes -- Comparing two groups : Time to event outcomes -- Comparing multiple factors -- Multi-level and hierarchical designs -- Ordinal data -- Dose-response studies.