Abstract
Seed germination experiments are conducted in a wide variety of biological disciplines. Numerous methods of analysing the resulting data have been proposed, most of which fall into three classes: intuition-based germination indexes, classical nonlinear regression analysis and time-to-event analysis (also known as survival analysis, failure-time analysis and reliability analysis). This paper briefly reviews all three of these classes, and argues that time-to-event analysis has important advantages over the other methods but has been underutilized to date. It also reviews in detail the types of time-to-event analysis that are most useful in analysing seed germination data with standard statistical software. These include non-parametric methods (life-table and Kaplan–Meier estimators, and various methods for comparing two or more groups of seeds) and semi-parametric methods (Cox proportional hazards model, which permits inclusion of categorical and quantitative covariates, and fixed and random effects). Each method is illustrated by applying it to a set of real germination data. Sample code for conducting these analyses with two standard statistical programs is also provided in the supplementary material available online (at http://journals.cambridge.org/). The methods of time-to-event analysis reviewed here can be applied to many other types of biological data, such as seedling emergence times, flowering times, development times for eggs or embryos, and organism lifetimes.
Keywords
failure-time analysis, frailty, Kaplan--Meier estimator, life-table estimator, log-rank test, reliability analysis, survival analysis
ScholarWorks Citation
McNair, James; Sunkara, Anusha; and Forbish, Daniel, "How to Analyse Seed Germination Data Using Statistical Timeto- event Analysis: Non-parametric and Semi-parametric Methods" (2013). Peer-reviewed scientific publications. 4.
https://scholarworks.gvsu.edu/peerscipub/4
