Instructor : Kenneth P. Burnham
Credits : 2 Semester Credits; letter (traditional) grading
Time : TBA
Place : TBA
Ref. #: TBA
Prerequisite: One statistics course, 300 level or higher for 2 credit hours (must have more statistics to take the additional hour of credit)
This will be a nonmathematical introduction to the philosophy of science and statistical planning and conduct of research studies (e.g., sampling studies, controlled experiments, quasiexperiments, monitoring studies, observational studies). We will not deal with statistical analysis formulae (e.g., ANOVA) for analyzing data. Rather, this course emphasizes issues of philosophy of science, valid study design, types of designs, and inference from studies. The overriding motivation for issues delved into here can be considered the simple question "how do we know what we (claim) to know."
A variety of concepts and issues will be studied; the following list is not intended to be xhaustive (note that stochastic variation & the necessity of using samples constitute the driving force underlying all this "statistical" stuff): variance components, samples vs. populations, inference from the sample to the population, nature of experimental proof, some philosophy of science, decisions vs. conclusions, scientific creativity, hypotheses null vs. alternative, Type I, II (and III) errors, significance levels, experimental units, controls, replication, randomization, pairing, matching of experimental units, blocking, stratification, covariates, balance, confounding factors, bias, blind and double blind protocols, randomized complete block design, incomplete block designs, Latin squares, crossover and switchback designs, multifactor designs, factorials, splitplot, splitblock, nested (hierarchal) designs, response surface, pre and posttreatment measurement designs, systematic designs, timesequential designs, measurement error, subsampling, pseudoreplication, repeated measures, nature of the treatments (nominal, ordinal, scaler), treatment effects and interactions, concepts of valid vs. invalid designs, assumptions, focused vs. diffuse research efforts, bias vs. precision, statistical "power," sample size, role of models in design and analysis.
There is no single text for this course; rather, there will be a substantial amount of studying and thinking required from a variety of sources, some entire books, some selected chapters and pages from books or papers from journals. The student is expected to do a lot of thinking, synthesizing and critical evaluation. Having memorization skills will not suffice to do well in this course; active participation is necessary. My goal is to impart to the student some fundamental principles, philosophy and knowledge of lasting value.
Your grade will be determined mostly by two interim exams and a final exam. An rare quiz is possible; there may be a small team project at the end of the course.
There are supposed to be 30 classroom meetings for this course, for 2 credit hours. This could be accomplished by meeting twice a week, or three times a week and then we do not meet every week of the semester. I need to do the latter: meet 3 times in many weeks. I cannot be here Sept. 1930. I will say more on this in class and on a separate page.
The course has the following general objectives:
The first two exams will each constitute about 1/4 of the grade; the final exam (comprehensive) will constitute about 1/2. Participation in class may matter. Class participation is criticalincluding reading the required material before class.
Extensive material will be made available
NoneDiscussion sessions will be held in class.
I will not attempt to help students design their specific thesis or dissertation projects as part of this class. The class will attempt to arm students with information that will help them in their own design for their graduate work. Their experimental design and subsequent analysis and inference should involve their Major Professor and Graduate Committee.
Always turn in your SSN (only) with exams, or other class material (never use your name); this facilitates fairness on my part in grading.
Here is a tentative course outline as regards specific lecture topics, and associated readings, or reading that I suggest be done in conjunction with these topics. I may alter the pace of presentation, or even the topics somewhat as I find necessary. Specific material to be studied is indicated in the list below as R or S denoting Required before that class, or Suggested to be read at about that time in the course.
The above will be revised, including adding two more classesprobably on finite population survey sampling. I expect the student to do a lot of reading (studying), and thinking about what they read, outside of class (something like 810 hrs. outofclass work per week). The following is a list of relevant references, some, but not all, of which will be utilized in this course. Shown with each reference are the pages that we are using, unless it is just a good background reference. Also coded is the importancevalue I place on the reference as regards the topics of this course: H (high), M (medium), L (low), or B coding a reference as background material of a type the student should be familiar with.
I would like to have found a single book that would be suitable as a text; classical statistical design books are not suitable for the purpose of this course. There are books that delve into issues of philosophy of science and design of research studies (e.g., experiments, quasiexperiments, sampling studies, observational studies, monitoring studies), in particular I found 3 such books that would make good texts for a course such as this one if the application area dealt with people; in order of there appropriateness (in my opinion) these books are: Cook and Campbell (1979) (heavy on sociological applications), Kish (1987) (heavy on survey sampling), and Anderson et al. (1980) (heavy on medical applications). If the student studied any one of these books from page 1 to the end, they could learn a lot. A fourth book in a similar vein is Cochran (1983), except Dr. Cochran died before finishing it. Mead (1988) is also a nice book, but too mathematical and too oriented towards agriculture for this course.
At the 1992 ESA meetings I found a book, just then published, written by Bryan Manly. As a text it would have some good features, but I would have to augment it with many topics and much other reading. Therefore, I am not going to use Manly (1992) as a text for the course. If you can afford the $36 cost of Manly (1992) you should buy it and use it.
Regarding philosophy of science books and their ilk, quite a few good ones exist. I suggest the student read all of Goldstein and Goldstein (1978) and Beveridge (1957). Also, Huff (1954) and Dethier (1962) are worth reading, both are light reading and somewhat humorous, but they make lots of good points. Two "heavier" reading books are Baker and Allen (1968), and Kuhn (1970); the latter is a classic in science. Somewhat lighter reading on philosophy of science are the two recent books, Ziman (1991) and Gillies (1993); I suggest you read (at least) parts of one of these. My preference is Gillies.
Assigned reading is marked with a leading * . Finally, I note that there is redundancy in these various assigned readings; this is deliberate.
Andelt, W.F. 1992. Effectiveness of livestock guarding dogs for reducing predation on domestic sheep. The wildlife Society Bulletin 20:5562. (L, but this is an example of a good retrospective quasiexperiment).
Anderson, R.L. 1990. Gertrude Mary Cox, January 13, 1990October 17, 1978. pp 117129 in Biographic Memoirs, National Academy of Sciences, Washington, D.C. (L, but you should know that much of statistics in the USA traces back to this remarkable woman Ms. Cox; this is paper is included in the packet, but is not an assigned reading)
*Anderson, S., A. Auquier, W.W. Hauck, D. Oakes, W. Vandaele, and H.I. Weisberg. 1980. Statistical Methods for Comparative Studies. John Wiley and Sons. New York. (15, 3237, 43: Section 4.6, 6971, & Chapter 14; all rated H)
Achen, C.H. 1986. The Statistical Analysis of Quasiexperiments. University of California Press. Berkeley, CA, USA. (L, entirely in a sociological setting, but very good on the differences of randomized vs. quasiexperimental designs and problems of the latter).
*Armour, C.L., K.P. Burnham and W.S. Platts. 1983. Field methods and statistical analyses for monitoring small salmonid streams. FWS/OBS/33. WELUT, USFWS, Ft. Collins, CO. (105127, H)
*Baker, J.J.W. and G.E. Allen. 1968. Hypothesis, Prediction, and Implication in Biology. AddisonWesley. Reading, MA. (3945, H) (a good general background book also.)
Barrow, J. D. 1991. Theories of everything: The quest for ultimate explanation. Clarenden Press, Oxford. 223 pp. (L for this class, but a good book re philosophy of science, see especially algorithmic compressibility).
*Bart, J. and J. D. Schoultz. 1984. Reliability of singing bird surveys: changes in observer efficiency with avian density. The Auk 101:307318. (H)
Bartholomew, G. A. 1982. Scientific innovation and creativity: A zoologist's point of view. American Zoologist 22:227235. (L)
*Beveridge, W.I.B. 1957. The Art of Scientific Investigation. Norton. New York. (Chapters 5 [except the middle of page 86] & 6 H; the entire book is worth reading, it is a classic. It is available in paperback for around $5)
Box, G.E.P. 1974. Statistics and the environment. J. Washington Academy of Science 64:5259. (L)
Box. G.E.P. 1976. Science and statistics. J. of the American Statistical Association 71:791799. (L)
*Box, G.E.P., W.G. Hunter and J.S. Hunter. 1978. Statistics for Experimenters: An Introduction to Design, Data Analysis and Model Building. John Wiley & Sons. New York. (117, H)
Box, Joan. Fisher. 1978. R. A. Fisher: The Life of a Scientist. John Wiley & Sons. New York. (B)
*Box, Joan Fisher. 1980. R. A. Fisher and the design of experiments, 19221926. The American Statistician 34(1):17. (H)
*Campbell, D.T. 1972. Measuring the effects of social innovations by means of time series. Pp 120129 in Statistics, A Guide to the Unknown, J.M. Tanur, F. Mosteller, W.H. Kruskal, R.F. Link, R.S. Pieters, and G.R. Rising, editors. HoldenDay, Inc., San Francisco. (H)
*Carpenter, S. R. 1990. Largescale perturbations: opportunities for innovation. Ecology 71(6):20382043. (M) Carpenter, S. R., T. M. Frost, D. Heisey and T. K. Kratz. 1989. Randomized intervention analysis and the interpretation of wholeecosystem experiments. Ecology 70(4):11421152 (M).
Carpenter, S. R. and P. A. Matson. 1990. Special feature: Statistical analysis of ecological response to largescale perturbations. Ecology 71(6):2037. (H)
Casti, J. L. 1989. Alternate Realities: Mathematical Models of Nature and Man. John Wiley and Sons. New York. 493 pages (Chapter 9 is "How do we know?: myths, models and paradigms in the creation of beliefs." It is good, but abstract, M)
*Chamberlin, T.C. 1965. The method of multiple working hypotheses. Science 148:754759. (H)
Chalmers, A. 1990. Science and its fabrication. University ofMinnesota Press, Minnesota, MN. (L)
*Cochran, W.G. 1976. Early development of techniques in comparative experimentation. Pages 126 in On the History of Statistics and Probability, D.B. Owen, editor. Marcel Dekker, Inc., New York. (H)
*Cochran, W.G. 1983. Planning and Analysis of Observational Studies. John Wiley & sons. New York. (130139, H)
Cochran, W.G. and G.M. Cox. 1950. Experimental Design. John Wiley, New York. (page 299, splitplot example H)
*Cook, T.D. and D.T. Campbell. 1979. Quasiexperimention: Design and Analysis Issues for Field Settings. Houghton Mifflin Co. Boston. (19 H and 341354 H).
Cormack, R.M. 1971. The Statistical Argument. Oliver and Boyd, Edinburgh. (B)
*Cox, D.R. 1958. Planning of Experiments. John Wiley & Sons, Inc. New York (pages 91108 M)
*Crowl, T. A. and A. P. Covich. 1990 Predatorinduced lifehistory shifts in a freshwater snail. Science 247:949951. (H)
Desu, M. M. and D. Roghavarao. 1991. Sample size methodology. Academic Press, Inc. New York. 135pp. (B)
Dethier, V.G. 1962. To Know a Fly. HoldenDay. San Francisco. (L, available in paperback)
Diamond, J. 1986. Overview: Laboratory experiments, field experiments, and nature experiments. Pp 322 in Community Ecology, J. Diamond and T.J. Case (eds.).Harper and Row, New York. (M)
*Dolby, G.R. 1982. The role of statistics in the methodology of the life sciences. Biometrics 38:10691083. (M for philosophy of science and statistics methods)
Eberhardt, L.L. 1978. Appraising variability in population studies. JWM 42:20738. (L)
*Eberhardt, L. L. and J. M. Thomas. 1991. Designing environmental field studies. Ecological Monographs 61(1):5373. (M)
Ehrenberg, A.S.C. and J.A. Bound. 1993. Predictability and prediction. Journal of the Royal Statistical Society, Series A 156,part 2:167206. (H, but I will not assign or use it).
Fagerstro m, T. 1987. On theory, data and mathematics in ecology. Oikos 50:258261. (L)
*Federer, W.T. 1983. Techniques useful in teaching a first course on statistical design. Pp 15 in Proceedings of the American Statistical Association. Section on Statistical Education. ASA, Washington D.C. (page 1 only, M)
Fienberg, S.F. and J.M. Tanur. 1989. Combining cognitive and statistical approaches to survey design. Science 243(24 Feb. 1989):10171023. (L).
*Fisher, R.A. 1960. The Design of Experiments (7th edition). Oliver and Boyd, London. (1125, H)
Fowler, N. 1990. The 10 most common statistical errors. Bulletin of the Ecological Society of America 71(3):161164. (L)
Gavin, T.A. 1989. What's wrong with the questions we ask in wildlife research? Wildlife Society Bulletin 17:345350. (M)
Gerrodette, T. 1987. A power analysis for detecting trends. Ecology 68(5):13641372 (H).
*Gessaman J.A. and K.A. Nagy. 1988. Transmitter loads affect the flight speed and metabolism of homing pigeons. The condor 90:662668. (H)
Gill, R.B. 1985. Wildlife researchan endangered species. Wildlife Society Bulletin 13:580587. (M)
Gillies, D. 1993. Philosophy of Science in the Twentieth Century. Blackwell, Oxford, UK. 251 pp. (H)
Goldstein, M. and I.F. Goldstein. 1978. How we Know. Plenum Press. New York. (the whole book, H)
Green, R.H. 1979. Sampling Design and Statistical Methods for Environmental Biologists. John Wiley & Sons, New York. (front and back covers, & 5456, 161218, L)
Green, R. H. 1989. Power analysis and practical strategies for environmental monitoring. Environmental Research 50:195205. (M)
*Gruenberger, F.J. 1954. A measure for crackpots. Science 145:14131415. (H)
*Gustavson, C. R., D. J. Kelly and M. Sweeney. 1976. Preylithium aversions. I: coyotes and wolves. Behavioral Biology 17:6172. (H, as an example of "bad")
Hahn, G. J. 1984. Experimental design in the complex world. Technometrics 26(1):1931. (L)
Hahn, G. J. and W. Q. Meeker. 1993. Assumptions for Statistical Inference. The American Statistician 47(1):111.
Hairston, N. G. 1989. Ecological Experiments: Purpose, design and execution. Cambridge University Press, Cambridge, England (M).
Hargrove, W.W. and J. Pickering. 1992. Psuedoreplication: a sine qua non for regional ecology. Landscape Ecology 6:251258. (H for issues of largescale ecology, experimentation and replication).
*Harshbarger, B. 1974. An example of how a designed experiment saved research workers from false recommendations. The American Statistician 28(4):128129. (H)
*Hayne, D.W. 1978. Experimental designs and statistical analyses in small mammal population studies. Pages 313 in Populations of Small Mammals Under Natural Conditions, D.P. Snyder, editor. Volume 5, Special Publication Series, Pymatuning Laboratory of Ecology, University of Pittsburgh, Linesville, PA. (H)
Heath, O.V.S. 1970. Investigation by Experimentation. Studies in Biology #23. The Institute of Biology. Arnold Ltd., London. (B)
*Heisterberg, J.F. and D.L. Otis. 1983. A changeover test design to compare relative efficacies of bird repellent seed corn treatments. Pages 98109 in Vertebrate Pest Control and Management Materials: Fourth Symposium, ASTM STP 817, D.E. Kaukeinen, editor. American Society for testing materials, Philadelphia. (H)
*Horgan, J. 1991. Profile: Reluctant revolutionary. Scientific American, (May 1991):4849. (H)
Howard, G. S. 1985. Basic Research Methods in the Social Sciences. Scott, Foresman and Co., Glenview, Illinois. 249pp+appendices. (L)
Huck. S. W., W. H. Cormier and W. G. Bounds. 1972. Reading Statistics and Research. Harper & Row. New York. (This books is intended for " people who do not engage in the research process but are interested in finding out about the results of others' investigations, " [page 1]. Part three is on design. The entire book is oriented to nonmedical human research sociology and psychology. However, the design part could be useful reading). (L)
Hurlbert, S.H. 1984. Pseudoreplication and the design of ecological field experiments. Ecological Monographs 54(2):187211. (M)
Huff, D. 1954. How to Lie with Statistics. W.W. Norton & Co, Inc. New York. (H, another classic. Available in paperback for around $5)
Jaeger, R. M. 1990. Statistics: A Spectator Sport, Second Edition. Sage Publications, Newbury Park, CA. 402 pp. (H: Chapter 6, i.e. pages 103134). (H, very good, but is entirely about sociological applications)
Jasby, A. D. and T. M. Powell. 1990. Detecting changes in ecological time series. Ecology 71(6):20442052. (M)
*Jeffers, J. N. R. 1978. Design of Experiments. Institute of Terrestrial Ecology. England. (M)
*Joiner, B.L. 1981. Lurking variables: some examples. The American Statistician 35(4):227233. (H)
Jones, B. and M.G. Kenward. 1989. Design and analysis of crossover trails. Chapman and Hall. London. (B)
Kamil, A.C. 1988. Experimental Design in Ornithology. Pp 313346 in Current Ornithology, Vol. 5, R.F. Johnston (ed.). Plenum Press, New York. (M)
Keith, L. H. (ed.). 1988. Principles of Environmental Sampling. American Chemical Society. Washington, DC (L, deals with sampling for chemical contamination, mostly of abiotic environment; it might be useful to an ecotoxicologist).
*Kendall, M.G. 1954. Hiawatha designs an experiment. The American Statistician 13:2324. (H)
*Kish, L. 1987. Statistical Design for Research. John Wiley and Sons. New York. (242247, H)
Kuhn, T.S. 1970. The Structure of Scientific Revolutions (second edition). University of Chicago Press, Chicago. (B, available in paperback; not expensive)
Lehmann, E.L. 1993. The Fisher, NeymanPearson theories of testing hypotheses: one theory or two? Journal of the American Statistical Association 88:12421249. (L)
Lehner, P.N. 1979. Handbook of ethological methods. Garland STPM Press, New York.403pp. (L for general use, but H for persons interested in ethology)
Likens, G. E. (Ed.). 1989. LongTerm Studies in Ecology: Approaches and Alternatives. SpringerVerlag. New York. (H value re design aspects of long term studies and some philosophy of different study types, esp. monitoring vs. experiments)
McAllister, M.K. and R.M. Peterman. 1992. Experimental design in the management of fisheries: a review. North American Journal of Fisheries Management 12:118. (M)
*MacNab, J. 1983. Wildlife management as scientific experimentation. Wildlife Society Bulletin 11:397401. (H)
Manly, B.F.J. 1992. The design and analysis of research studies. Cambridge University Press, Cambridge, UK 353pp. (parts of this book I rate H as regards this course).
Manly, B.F.J. and V.L. Wright. 1982. Report of workshop on design of environmental impact statements. Special Report 8201, Department of Experimental Statistics, LSU, Baton Rouge, Louisiana. (L)
Mantel, N. 1976. A personal perspective on statistical techniques for quasiexperiments. Pages 103129 in On the History of Statistics and Probability, D.B. Owen, editor. Marcel Dekker, Inc., New York. (L)
Maxwell, S. E. and H. D. Delaney. 1990. Designing experiments and analyzing data: A model comparison perspective. Wadsworth Publishing Company. Belmont, California. (Part 1, Chapter 1 (M) and Chapter 2 (H)).
Mead, Roger. 1988. The design of experiments: statistical principles for practical applications. Cambridge University Press, New York. (good, but oriented to agriculture, hardback costs $130, paperback is $37.50). (L)
Mead, R. 1990. The nonorthogonal design of experiments. Journal of the Royal Statistical Society, series A, vol 153(2):151201. (L)
Mead, R., and R.N. Curnow. 1983. Statistical Methods in Agriculture and Experimental Biology. Chapman and Hall. New York. (Chapter 14, which is pages 284309 M)
*Neufeld, A. H. 1989. Reproducing results (letter to editor). Science. (H)
Nudds, T.D. and M.L. Morrison. 1991. Ten years after "Reliable Knowledge": are we gaining? JWM 55:757760. (H, included in the reading packet)
*Ostle, B. 1963. Statistics in Research. Iowa State Univ. Press, Ames, Iowa. (244256, 271273, M)
Ostle, B. and L.C. Malone. 1987. Statistics in Research: Basic Concepts and Techniques for Research Workers (4th edition). Iowa State Univ. Press, Ames, Iowa. (B)
*Payne, S.L. 1951. The Art of Asking Questions. Princeton University Press. Princeton , New Jersey. (Foreword, 316 H)
Pease, C.M. and J.J. Bull. 1992. Is science logical? BioScience 42:293298. (M)
Peterman, R. M. 1990. Statistical power analysis can improve fisheries research and management. Canadian Journal of Fisheries and Aquatic Science 47:215 (M)
*Platt, J.R. 1964. Strong inference. Science 146:347353 (H)
Quinn, J.F., and A.E. Dunham. 1983. On hypothesis testing in ecology and evolution. American Naturalist 122(5):2237 (L)
Pease, C. M. and J. J. Bull. 1992. Thinking of Biology: Is science logical? BioScience 42:293298. (L)
*Rayner, A.A. 1986. Some sidelights on experimental design. Pp 245266 in The Fascination of Statistics, R.J. Brook, G.C. Arnold, T.H. Hassard and R.M. Pringle (eds.). Marcel Dekker, Inc. New York. (H)
Reckhow, K. H. 1990. Bayesian inference in nonreplicated ecological studies. Ecology 71(6):20532059. (L).
Ribeyre, F. 1985. Problems and methodologies in ecotoxicology: biological models and experimental design. Ecotoxicology and Environmental Safety 9:346363 (L).
Rinne, J.N. 1988. Grazing effects on stream habitat and fishes: research design considerations. North American Journal of Fisheries Management 8:240247. (M)
*Romesburg, H.C. 1981. Wildlife science: gaining reliable knowledge. JWM 45:293313 (H)
Romesburg, H.C. 1991. On improving the natural resources and environmental sciences. JWM 55:744756. (H, included in the reading packet)
*RootBernstein, R. S. 1989. How scientists really think. Perspectives in Biology and Medicine 32(4):472488. (Long, but very well written; deals with philosophy of science issues) (H)
Rose, K. A. , and E. P. Smith. 1992. Experimental design: the neglegcted aspect of environmental monitoring. Environmental Management 16:691700. (L)
Rosenbaum, P. R. 1995. Observational Studies. SpringerVerlag, New York, NY. 230pp (L for the book as a whole because it focuses too much on analysis; H for parts on design considerations).
Salmon, S. L. and A. A. Hanson. 1964. The Principles and Practice of Agricultural Research. Leonard Hill, London. (77111, 112127, 128140, L)
Scheiner, S. M. and J. Gurevitch (Eds.). 1993. Design and Analysis of Ecological Experiments. Chapman and Hall, London. (used pages 6266 on the split plot design, otherwise M).
Schumacher, P. and J. V. Zidek. 1993. Using prior information in designing intervention detection experiments. Annals of Statistics 21:447463 (not for this course).
Stuart, Alan. 1984. The Ideas of Sampling (3 rd edition). Oxford University Press, New York. 91pp (a good book for an introduction to the ideas of sampling finite populations). (B)
*Tacha, C.T., W.D. Warde and K.P. Burnham. 1982. Use and interpretation of statistics in wildlife journals. Wildlife Society Bulletin 10:355362. (M)
Tukey, J. W. 1960. Conclusions vs. decisions. Technometrics 2:423433. (L)
Underwood, A. J. 1990 Experiments in ecology and management: their logics, functions and interpretations. Australian Journal of Ecology 15:365389. (M)
Yoccoz, N. G. 1991. Commentary: Use, overuse, and misuse of significance tests in evolutionary biology and ecology. Bulletin of the Ecological Society of America 72:106111. (L, re design issues, H otherwise)
Walters, C. J., J.S. Collie, and T. Webb. 1988. Experimental designs for estimating transient responses to management disturbances. Canadian Journal of Fisheries and Aquatic Science 45:530538. (L, deals with stairstep designs, aka staggered entry).
Walters, C. J. and C. S. Holling. 1990. Largescale management experiments and learning by doing. Ecology 71(6):20602068. (M)
Wang, C. 1993. Sense and Nonsense of Statistical Inference: Controversy, Misuse, and Subtlety. Marcel Dekker, Inc, New York. 244 pp. (M)
Webster, D. B. 1992. Viewpoint: Replication, randomization, and statistics in range research. Journal of Range Management 45:285290. (M)
Wiens, J. A. 1989. The ecology of bird communities, Volume 1: foundations and patterns. Cambridge University Press, New York, NY. 539pp. (Part I, which is pages 1 to 70, M for this course, but excellent reading if you can make the time).
Wilson, M.F. 1981. Commentary: Ecology and Science. Bulletin of the Ecological Society of America 62:412. (philosophical, M)
Van Latesteijn, H.C. and R.H.D. Lambeck. 1986. The Analysis of Monitoring data with the aid of timeseries analysis. Environmental monitoring and assessment 7:287297. (L)
Ziman, J. 1991. Reliable Knowledge: An Exploration of the Grounds for Belief in Science. Cambridge University Press, Cambridge, UK. 197 pp. (M).