Course Objective: The purpose of this class is to give students experience in real-world sampling design and data analyses. Students will be required to develop experimental designs for a variety of topics and conduct analyses using real-world data. Each topic will be presented with data and examples.
Format: This course will be a mixture of lecture, computer lab exercises, and discussion. Generally we will have lecture on Fridays, followed by a lab session on Monday where students will analyze data based on the material presented in the lecture. This will be followed by a discussion on Wednesday focusing on classic and contemporary issues in designing and analyzing data from field experiments.
Instructors: This course will be team taught through faculty associated with the Program for Environmental Statistics (PES). There are four main instructors, as well as several guest lecturers.
Department of Fisheries and Aquatic Sciences
Campus office building 851
(352) 392 9618 ext 252; email: msal@ufl.edu
Office hours: Monday 8:30-9:30, or by appointment
School of Forest Resources & Conservation
349 Newins-Ziegler Hall
(352) 846-3503; email: staudham@ufl.edu
Office hours: Monday 1-3, or by
appointment
Department of Fisheries and Aquatic Sciences
Campus office building 851
(352) 392 9618 ext 270; email: billpine@ufl.edu
Office hours: Monday 8:30-9:30, or by appointment
Department of Statistics
IFAS Statistics Consulting Unit
406 McCarty Hall C
(352) 392 3724; email: mcxman@ufl.edu
Office hours: Tuesday, 10:45-11:45 or by appointment
Course Pre-requisite: STAT 6166 or equivalent, or approval of instructors
Computers and software: There will be extensive use of computers inside and outside of class. Examples and assignments will be given in R, Excel, and/or SAS. Students may choose the platform which best fits their needs; however, programming support may not always be possible in each platform.
Grades: There will be 5 assignments based on analysis techniques presented in this course (10 points each). There will also be two written assignments (20 points each): (1) a critique of the design and analysis approaches from two papers in your field, (2) a detailed research proposal that synthesizes several key aspects of this class (e.g., clearly defined, testable hypothesis; justification for experimental design; planning support based on simulated data or power analysis; examples from the literature; etc.). Because discussion and interaction is a key component of the Wednesday class, your participation will be worth 10 points. To receive full participation, students should regularly demonstrate (via informed comments or questions) that they have read the required papers and are completing the lab exercises.
Listserve:
listserv@lists.ufl.edu
In the message type
subscribe fall-3973-l YOUR NAME
And insert your name where it says your name. Make sure to remove any signature from the bottom of your email. The server will recognize the email account that you sent the message from, and add that account to the list. After you subscribe, you will receive a message that explains the rules of the list. If you have any problems, follow this link
http://docweb.cns.ufl.edu/docs/d0099/d0099.html
References: Here is a partial list of reference books that we use frequently
Tentative Schedule (subject to change)
|
Week |
Date |
Monday: 10:40am – 12:40pm |
Wednesday: : 10:40am – 11:35am |
Friday: 10:40am – 12:40pm |
|
1 |
21-Aug |
|
Welcome and Introduction Info cards on stats and computer software experience |
Module 1: Why Ecol stats -designing “experiments” to inform policy makers/managers -Asking good questions -What is your elevator statement? BP |
|
2 |
28-Aug |
M1: Identifying stakeholders and developing testable hypotheses -intro and orientation to computer lab basic Excel/R/SAS -PLOT YOUR DATA MA |
Discuss M1 Readings: Krebs Ch1,.Otis 2001 -Students present their elevator statement ALL |
Module 2: Samp./Exp. Design -empirical vs. experimental approaches -overview of sampling designs CS |
|
3 |
4-Sep |
Labor Day - No Class |
M2: Samp./Exp. Design -wrap-up lecture from Fri Discuss M2 Readings: - Krebs Ch 8, 10 CS |
M2: Samp./Exp. Design -sample size & power analysis
MC |
|
4 |
11-Sep |
M2: Samp./Exp. Design -sample size & power analysis lab
BP |
Discuss M2 Readings: -Hurlbert (1984,) Oksanen (2001), Hurlbert (2004) M2: Discussion of experimental design and pseudoreplication ALL |
M2: Samp./Exp. Design -Continue discussion from Wednesday -Ecosystem experiments BP |
|
5 |
18-Sep |
M2: Samp./Exp. Design -basic stats crash course -into to SAS **NOTE ROOM CHANGE TO ARCH 116 (Architecture Building) CS |
M2: Samp./Exp. Design -Carpenter (1990), Walters and Holling 1990 -Discuss lab exercises ALL |
Module 3: Basic Data Analysis -Applied correlation and regression
MA |
|
6 |
25-Sep |
M3: Basic Data Analysis -Regression and correlation lab
MA |
M3: Basic Data Analysis -101 ways to use regression
ALL |
Module 4: Data Analysis -assumptions -basic one way, two way factorial, RCB -interpretation of interactions -MCP’s MA |
|
7 |
2-Oct |
M4:
Data Analysis -ANOVA -Testing assumptions -Multiple comparison procedures MA |
M4: Data Analysis -Why statistics is made up? |
Homecoming - no class |
|
8 |
9-Oct |
M4: Data Analysis -ANCOVA MC |
M4: Data Analysis -Dependent data -WRITING ASSIGNMENT 1 DUE |
Faculty administrative conflict – no class |
|
9 |
16-Oct |
M4: Data Analysis -Split Plot ANOVA exercise CS |
M5: Are the designs we discussed applicable to non-normal data? |
CS |
|
10 |
23-Oct |
M5: Data Analysis -Repeated measures exercise CS and MA |
M5: Data Analysis -papers on repeated measures -time dependent |
M5: A bit on nonparametric stats and non-linear model fitting
BP |
|
11 |
30-Oct |
M5: Data Analysis -non-linear regression exercise BP |
M5: Data Analysis "My data doesn't look like that" Discussion on where to start with messy data |
M6:Indices
what are they and how are they used and misused? -diversity -competition -habitat -IBI BP |
|
12 |
6-Nov |
M5: Data Analysis -Indicies lab MA |
M5: Data Analysis -Introduction to randomization, bootstrapping and MC test methods MC |
Holiday – no class |
|
13 |
13-Nov |
M5: Data Analysis -Bootstrapping/Monte Carlo sampling MC |
M5: Data Analysis -advantages/disadvantages of non-parametric methods MC |
M5: Data Analysis - Bayes’ theorem and method -Likelihood functions and tests Dr. Bob Dorzaio, USGS and UF PES |
|
14 |
20-Nov |
Module 5: Permutation testing... MC |
M 6: Tentative- class cancelled |
Thanksgiving – no class
|
|
15 |
27-Nov |
Tentative: Adaptive management lecture by Carl Walters and Buzz Holling (still trying to finalize) |
Module 6: Detection probability issues.... BP |
Module 6: Occupancy Models Special Lecture: Dr. Bob Dorzaio, USGS and UF PES "A multispecies site-occupancy model for estimating species richness and other attributes of biological communities" WRITING ASSIGNMENT 2 DUE |
|
16 |
4-Dec |
Module 8: |
M8: Review and evaluations END OF CLASS
|
|
Update 10/23
Labs: You must chose two labs from the following to turn in (the labs will be due 1 week from their assignment date): repeated measures, nonlinear, indicies, or bootstrap.
Schedule: The schedule is subject to change as we are working to bring in a few outside speakers
Writing assignments: I will return the first writing assignment this week. The second writing assignment is posted above and is due December 1. Late writing assignments will not be accepted.
Some References
Anderson, D. R., K. P. Burnham, and W. L. Thompson. 2000. Null hypothesis testing: problems, prevelance, and an alternative. Journal of Wildlife Management 2000:912-922.
Anderson, D. R., K. P. Burnham, W. R. Gould, and S. Cherry. 2001. Concerns about finding effects that are spurious. Biometrics 29:311-316.
Carpenter et al 1995 Science 269 324-327. Ecological expts with Model Systems.
Carpenter, S. R. 1990. Large-scale perturbations: Opportunities for innovation. Ecology 71:2038-2043.
Efron, B., and R. Tibshirani. 1991. Statistical data analysis in the computer age. Science 253:390-395.
Hatfield, J.H., W.R. Gould, B.A. Hoover, M.R. Fuller, E.L.Lindquist. 1996. Detecting trends in raptor counts: power and type I error rates of various statistical tests. Wildlife Society Bulletin 24:505-515.
Hurlbert, S. H. 1984. Pseudoreplication and the design of ecological experiments. Ecological Monographs 54, 187- 211.
Hurlbert, S. H. 2004. On misinterpretations of pseudoreplication and related mattes: a reply to Oksanen. Oikos 104:591-597.
Kendall, W. L., and W. R. Gould. 2002. An appeal to undergraduate wildlife programs: send scientists to learn statistics. Wildlife Society Bulletin 30:623-627.
Legg, C.J.; Nagy, L. 2006. Why most conservation monitoring is, but need not be, a waste of time. J. Environ. Manage. 78(2):194-199
Levin, S. 1992. The problem of pattern and scale in Ecology. Ecology 73: 1943-1967.
McAllister, M.K. and R.M. Peterman. 1992. Experimental design in the management of fisheries: a review. N. Amer. J. Fish. Management 12:1-18.
Oksanen, L. 2001. Logic of experiments in ecology: is pseudoreplicatoin a pseudoissue? Oikos 94:27-38.
Otis, D. L. 2001. Quantitative training of wildlife graduate students. Wildlife Society Bulletin: 29 1043-1048.
Peterman, R.M. 1990. Statistical power analysis can improve fisheries research and management. Can. J. Fish. Aquat. Sci. 47:2-15.
Walters, C. J. and C. S. Holling. 1990. Large-scale management experiments and learning by doing. Ecology 71:2060-2068.