Bio 621 - Readings in Biostatistics

New: Project Description

Readings:
Week: 1 2 3 4 5 6 7 8 9 10 11 12

Notes:
Week: 0 1 2

Readings

Week 0: Introduction, On the Tendencies of Motion, Strong Inference Approach

  1. Nabi, I. (1985). On the tendencies of motion. In The Dialectical Biologist, Cambridge: Harvard University Press.[handout in class]
  2. Platt, J. R. (1964). Strong Inference. Science 146, 347-353.
  3. O'Donohue, W. and Buchanan, J. A. (2001). The weaknesses of strong inference. Behavior and Philosophy 29, 1-20.
  4. Quinn, J. F. and Dunham, A. E. (1983). On hypothesis testing in ecology and evolution. Am. Nat. 122, 602-617.
  5. Loehle, C. (1987). Hypothesis testing in ecology: psychological aspects and the importance of theory maturation. Quarterly Review of Biology 62, 397-409.
  6. Murray, B. G., Jr. (2004). Laws, hypotheses, guesses. American Biology Teacher 66, 598-599.
  7. Sit, V. (1998). On the presentation of statistical results: a synthesis. In Biometrics Information.

Week 1: alpha, beta, Power

required

  1. Thomas, L. and Juanes, F. (1996). The importance of statistical power analysis: an example from Animal Behavior. Animal Behaviour 52, 856-859.
  2. Hoenig, J. M. and Heisey, D. M. (2001). The abuse of power: the pervasive fallacy of power calculations for data analysis. American Statistician 55, 19-24.
  3. Stoehr, A. M. (1999). Are significance thresholds appropriate for the study of animal behavior? Animal Behaviour 57, F22-F25.
  4. Di Stefano, J. (2001). Power analysis and sustainable forest management. Forest Ecology and Management 154, 141-153.

Quick notes - Several authors, such as Thomas and Juanes, and even journals are now advocating post hoc power analyses to determine if the failure to reject the null hypothesis is due to (1) the null hypothesis being true, or (2) the alternatve hypothesis is actually true but the experiment did not have enough data (power) to reject the null hypothesis (this is a type II error). The problem is that if a stasticial test fails to reject the null hypothesis, there is by definition not enough power because of the 1-to-1 relationship between P-values and power! Hoenig shows that a power analysis can never test the null hypothesis but offers an alternative called equivalence testing. So post-hoc power analysis can really only tell you that you should have measured more data! But then this begs the question, is the paper publishable if I didn't measure enough data to test the hypothesis? Ouch! Since one can probably always find significance given enough data, this also begs the question, Is there any sense to hypothesis testing? I would say yes but only if you have some a priori notion of what the biologically minimum effect size should be because if you can only find signficance at an effect size below this, then the effect is biologically trivial. Di Stefano is a good example of how to control for type II error by not only measuring more samples, but by also adjusting alpha, although this will depend on the costs of each type of error. Stoehr nicely argues that costs of type I or type II errors are not relevant to many purely academic (i.e. not applied) questions addressed by biologists and advocates emphasizing effect size and not treating the P-value as an either-or test.

Week 2: Pseudoreplication

Required

  1. Hurlbert, S. H. (1984). Pseudoreplication and the design of ecological field experiments. Ecological Monographs 54, 187-211.

Suggested

  1. Oksanen, L. (2001). Logic of experiments in ecology: is pseudoreplication a pseudoissue? Oikos 94, 27-38.
  2. Heffner, R. A., Butler, M. J., IV and Reilly, C. K. (1996). Pseudoreplication revisited. Ecology 77, 2558-2562.
  3. Cottenie, K. and De Meester, L. (2003). Comment to Oksanen (2001): reconciling Oksanen (2001) and Hurlbert (1984). Oikos 100, 394-396.
  4. Hurlbert, S. H. (2004). On misinterpretations of pseudoreplication and related matters: a reply to Oksanen. Oikos 104, 591-597.
  5. Jenkins, S. H. (2002). Data pooling and type I errors: a comment on Leger & Didrichsons. Animal Behaviour 63, F9-F11.
  6. Garland, T., Jr. and Adolph, S. C. (1994). Why not to do two-species comparative studies: limitations on inferring adaptation. Physiological Zoology 67, 797-828.

Week 3: Directional tests. Post hoc tests.

Required

  1. Rice, W. R. and Gaines, S. D. (1994). 'Heads I win, tails you lose': testing directional alternative hypotheses in ecological and evolutionary research. Trends in Ecology and Evolution 9, 235-237.
  2. Day, R. W. and Quinn, G. P. (1989). Comparisons of treatments after an analysis of variance in ecology. Ecological Monographs 59, 433-463.
  3. Rice, W. R. (1989). Analyzing tables of statistical tests. Evolution 43, 223-225.

Suggested

  1. Gaines, S. D. and Rice, W. R. (1990). Analysis of biological data when there are ordered expectations. American Naturalist 135, 310-317.

Initial Notes. These are heavy hitters. The Day and Quinn 1989 has over 1100 citations while the Rice 1989 paper has over 4400 citations! That is a really, really phenomenal number. How many papers in ecology, evolution, and related fields (that would read the journal Evolution) have even been published since 1989? We talked about the sequential Bonferroni in the Rice 1989 paper in Bio 601 but we will spend much more time on it here!


Week 4: Meta-analysis

Required

  1. Arnqvist, G. and Wooster, D. (1995). Meta-analysis: synthesizing research findings in ecology and evolution. Trends in Ecology and Evolution 10, 236-240.
  2. Osenberg, C. W., Sarnelle, O., Cooper, S. D., et al. (1999). Resolving ecological questions through meta-analysis: goals, metrics, and models. Ecology 80, 1105-1117.
  3. Treseder, K. K. (2004). A meta-analysis of mycorrhizal responses to nitrogen, phosphorus, and atmospheric CO2 in field studies. New Phytologist 164, 347-355.

Suggested

  1. Gates, S. (2002). Review of methodology of quantitative reviews using meta-analysis in ecology. Journal of Animal Ecology 71, 547-557.

Initial notes. Notice the Arnqvist paper is another TREE paper that you can use as a model for your own paper. The Treseder paper is simply an example of meta-analysis. Don't bother to read it in detail. Osenberg et al. show how the measure of the effect is model dependent.

Week 5: Bootstrap, randomization, and Monte-Carlo methods

Required
  1. Crowley, P. H. (1992). Resampling methods for computation-intensive data analysis in ecology and evolution. Annual Review of Ecology and Systematics 23, 405-447.

Week 6: Model II regression

Required

  1. Ricker, W. E. (1984). Computation and uses of central trend lines. Canadian Journal of Zoology 62, 1897-1905. [handout in office]

Week 7: Multiple regression and path analysis

Required

  1. Kingsolver, J. G. and Schemske, D. W. (1991). Path analyses of selection. Trends in Ecology and Evolution 6, 276-280.
  2. Petraitis, P. S., Dunham, A. E. and Niewiarowski, P. H. (1996). Inferring multiple causality: the limitations of path analysis. Functional Ecology 10, 421-431.

Suggested

  1. Grace, J. B. and Pugesek, B. H. (1998). On the use of path analysis and related procedures fro the investigation of ecological problems. American Naturalist 152, 151-159.
  2. Mitchell, R. J. (1992). Testing evolutionary and ecological hypotheses using path analysis and structural equation modelling. Functional Ecology 6, 123-129.
  3. Smith, F. A., Brown, J. H. and Valone, T. J. (1997). Path analysis: a critical evaluation using long-term experimental data. American Naturalist 149, 29-42.
  4. Sokal, R. R. and Rohlf, F. J. (1994). Biometry. San Francisco: W. H. Freeman. Handout in office.
  5. Wright, S. (1918). On the nature of size factors. Genetics 3, 367-374.
  6. Wright, S. (1921). Correlation and causation. Journal of Agricultural Research 20, 557-585.
  7. Wright, S. (1932). General, group and special size factors. Genetics 17, 603-619.

Week 8: Ratios, size adjustment, ANCOVA

Required

  1. Packard, G. C. and Boardman, T. J. (1999). The use of percentages and size-specific indices to normalize physiological data fro variation in body size: wasted time, wasted effort? Comparative Biochemistry and Physiology 122, 37-44.
  2. Berges, J. A. (1997). Ratios, regression statistics, and "spurious" correlations. Limnology and Oceanography 42, 1006-1007.
  3. Liermann, M., Steel, A., Rosing, M., et al. (2004). Random denominators and the analysis of ratio data. Environmental and Ecological Statistics 11, 55-71.

Suggested

  1. Atchley, W. R., Gaskins, C. T. and Anderson, D. (1976). Statistical properties of ratios. I. empirical results. Systematic Zoology 25, 137-148.
  2. Atchley, W. R. and Anderson, D. (1978). Ratios and the statistical analysis of biological data. Systematic Zoology 27, 71-78.
  3. Beaupre, S. J. and Dunham, A. E. (1995). A comparison of ratio-based and covariance analyses of a nutritional data set. Functional Ecology 9, 876-880.
  4. Garcia-Berthou, E. (2001). On the misuse of residuals in ecology: testing regression residuals vs. the analysis of covariance. Journal of Animal Ecology 70, 708-711.
  5. Jasienski, M. and Bazzaz, F. A. (1999). The fallacy of ratios and the testability of models in biology. Oikos 84, 321-326. [Handout in office]
  6. Spurrier, J. D., Hewett, J. E. and Lababidi, Z. (1982). Comparison of two regression lines over a finite interval. Biometrics 38, 827-836.
  7. Tsutakawa, R. K. and Hewett, J. E. (1978). Comparison of two regression lines over a finite interval. Biometrics 34, 391-398.

Quick notes.The Atchley 1976 paper is a classic, but it is not required reading.But know it anyway just for your own career development. Also read the Atchley 1978 response to comments on Atchley 1976. Ouch! Bill Atchley was an undergraduate student of Jim Rohlf's at Kansas. His name will appear again later. The Spurrier 1982 and Tsutakawa 1978 papers are for ANCOVA problems when slopes differ between groups. I love these papers, but not enough people know about them. For numericophiles only!

Week 9: PC (phylogenetically correct) statistics

Required

  1. Ackerly, D. D. (1999). Comparative plant ecology and the role of phylogenetic information. In Physiological Plant Ecology, (ed. M. C. Press, J. D. Scholes and M. G. Barker), pp. 391-413. Oxford: Blackwell Scientific.
  2. Westoby, M., Leishman, M. and Lord, J. (1995). On misinterpreting the 'phylogenetic correction'. Journal of Ecology 83, 531-534.
  3. Harvey, P. H., Read, A. F. and Nee, S. (1995). Why ecologists need to be phylogenetically challenged. Journal of Ecology 83, 535-536.
  4. Westoby, M., Leishman, M. and Lord, J. (1995). Further remarks on phylogenetic correction. Journal of Ecology 83, 727-729.
  5. Harvey, P. H., Read, A. F. and Nee, S. (1995). Further remarks on the role of phylogeny in comparative ecology. Journal of Ecology 83, 733-734.

Suggested

  1. Felsenstein, J. (1985). Phylogenies and the comparative method. American Naturalist 125, 1-15.

Historical

  1. Clutton-Brock, T. H., Harvey, P. H. and Rudder, B. (1977). Sexual dimorphism, socionomic sex ratio and body weight in primates. Nature 269, 797-800.
  2. Ridley, M. (1983). The Explanation of Organic Diversity: the Comparative Method and Adaptations for Mating. Oxford: Oxford University Press.

Quick notes.This is a hugely important topic. Essentially modern, PC comparative methods attempt to account for the lack of independence in data when the "individuals" are species. This lack of independence is a consequence of the hierarchical relationship among species - some share a much more recent ancestor than others and because of this, we expect these more closely related species to share traits (morphological or ecological) simply because they inherited these from a common ancestor (along with some sort of stabilizing mechanism). Clutton-Brock and Harvey were the first to really identify the problem when using the comparative method to infer functional associations among traits and Ridley wrote a whole book on the problem and its relevance to the diversity of spider genitalia. But it was really Felsenstein's paper that shook up everyone.

Week 10 SC (spatially correct) statistics

Required

Perry, J. N., Liebhold, A. M., Rosenberg, M. S., et al. (2002). Illustrations and guidelines for selecting statistical methods for quantifying spatial pattern in ecological data. Ecography 25, 578-600.

Suggested

Dale, M. R. T., Dixon, P., Fortin, M.-J., et al. (2002). Conceptual and mathematical relationships among methods for spatial analysis. Ecography 25, 558-577.

Liebhold, A. M. and Gurevitch, J. (2002). Integrating the statistical analysis of spatial data in ecology. Ecography 25, 553-557.

Historically important

Sokal, R. R. and Oden, N. L. (1978). Spatial autocorrelation in biology 1. methodology. Biological Journal of the Linnaen Society 10, 199-228.

Quick notes. OK. The paper I would have picked is by Fortin and Legendre but its not in PDF format. The required and suggested papers are all from a theme issue of ecography, a journal that I had never heard of until getting these papers. So I haven't read these, yet, but the titles and abstracts look like they will be the best, most recent, introduction to the problems and methods.

Week 11 Multivariate statistics

  1. James, F. C. and McCulloch, C. E. (1990). Multivariate analysis in ecology and systematics: panacea or Pandora's box? Annual Review of Ecology and Systematics 21, 129-166.
  2. Campbell, N. A. and Atchley, W. R. (1981). The geometry of canonical variate analysis. Systematic Zoology 30, 268-280.

Quick notes. There is no way to discuss multivariate statistics in one 2-hour class so I will spend much of this time introducing PCA - principal components analysis, the understanding of which is critical to understanding anything else. We will then talk a little about the geometry of canonical variates analysis as described by Campbell and Atchley. And we will revisit x-transpose-x-inverse-x-transpose-y and how all the multivariate techniques are related to this equation.

Week 12 Model testing

  1. Johnson, J. B. and Omland, K. S. (2004). Model selection in ecology and evolution. Trends in Ecology and Evolution 19, 101-108.
  2. Anderson, D. R., Burnham, K. P. and Thompson, W. L. (2000). Null hypothesis testing: problems, prevalence, and an alternative. Journal of Wildlife Management 64, 912-923.
  3. Strong, D. R., Whipple, A. V., Child, A. L., et al. (1999). Model selection for a subterranean trophic cascade: root-feeding caterpillars and entomopathogenic nematodes. Ecology 80, 2750-2761.

Quick notes. Read the Anderson et al. paper up to the model testing section on Kullback-Leibler information, then switch to the Johnson and Omland paper. Read the Strong paper for an example.