This page contains material closely related to class material. This is provided if you want another perspective on lecture material.

• Weeks 1 and 2: principles and power analysis
  • General overview in wildlife, use of power analysis:
    Cherry 1998, Statistical tests in publications of The Wildlife Society
  • Perugini, Galluci, Costantini 2018. A practical primer to power analysis for simple experimental designs. International Review of Social Psychology, 31(1), 20. DOI: http://doi.org/10.5334/irsp.181. Covers a lot of designs and goes well beyond what we cover in 471/571. Includes G*Power and R code. Almost as good as a book.
  • A second overview of many sample size situations Dell et al 2012 paper. Most of the formula here are "quick" approximations. My approach deals more carefully with small sample sizes (the difference between Z quantiles and T quantiles).
  • Ramsey and Schaefer (Statistical Sleuth): section 23.5 describes the confidence interval width approach to determining sample size.

Week 3: subsampling and variance components

• Dixon and Ding, 2008. A metabolomic data uncertainty budget for the plant, Arabidopsis thaliana. Chance 20:12-17.
  This is a relatively complete exploration of sources and characteristics of variability in measurements of amino acid concentrations in a plant. The material on transformations is not relevant for us.

Week 5: blocking

• Dixon, 2016, Should blocks be fixed or random? Conference paper hosted on a newprairiepress open access site (https://doi.org/10.4148/2475-7772.1474)
• Doncaster et al 2014 Describes relative performance, which adds changes in df to the traditional relative efficiency measure.

Week 5: Another treatment of blocking

• Chapter 13, esp. sections 13.1, 13.2, and 13.3, pp 315-338 of Gary Oehlert's design text, now online at his UMN site