Kenneth H. Pollock
Deparments of Zoology (and Statistics)
231 David Clark Labs
515-3514
pollock@unity.ncsu.edu
Class: Tu/Th 5.45-7.00pm, DCL 102
OH: M 1.30-2.30, We 2.30-3.30 or By Appt
Ecological
Studies have very Varied Designs and Range from True Experiments
to Observational Studies . Two examples to show these extremes which we
discuss in the class are:
Ecological
Expts on Effects of Corridors (Nick Haddad)
Randomised Block Design with Block Structure Shown
Observational Studies on Birds (Ted Simons)
Large
Scale Field Studies based on Aural Point Counts on Neotropical Migrants
in Great Smokey Mountains National Park.
Many other examples will also be given.
| |
Elluminate Recordings Go to http://slms.delta.ncsu.edu, go to the date you want, Go to recordings and scroll though to find our lectures.
Lecture 1
Lecture 2
Lecture 3 Nick Haddad Guest
Corridors
Micocosms
Criticism of Micocosms
Lecture 4
Lecture 5
Lecture 6
Lecture 7
Hurlbert paper
Hefner Paper
Lecture 8
Likens Paper
Lecture 9 Pollock Component
Lecture 9 Tarr Component
Lecture 10
Lecture 11
Lecture 12-Jim Gilliam Guest
Jim Gilliam paper
Lecture 13
Lecture 14 Ted Simons Guest
Bird Radio Paper
Levin Paper
Lecture 15
Pollock et al. Sampling Paper
Nordstrom et al Mink Removal Paper
Lecture 16 Joe Hightower Guest
Spreadsheet
Lecture 17
Bird Song Paper (Nelson and Marler)
Predator/Prey Study (Scharf et. al.)
Lecture 18
Nueman and Wildman(2002)
Friedlander et al.(2003)
Lecture 19
Pomeroy Paper
Lecture 20
Shriner et al Paper
Lecture 21-Occupancy plus discuss the two papers below
Geist and Auerswald (2007)
Brown and Harris (2005)
Lecture 22 Long term Studies and Other Critiques
Brady and Slade (2004)
Waters (1999)
Means and Travis (2007)
Develey and Stouffer (2001)
Lecture 23 Modelling
Kareiva et al. (2000)
Lecture 24 Tom Wentworth Visit
Kenkel et al. (1989)
Lecture 25 Meta-Analysis Introduction
Take Home Exam
Lecture 26 Rob Dunn Visit -Meta Analysis
Dunn (2004)
Lecture 27 Conclude Meta-Analysis
Lecture 28 Final Lecture
Guest
Speakers
Jim Gilliam Feb 21
Ted Simons ( http://www4.ncsu.edu/~simons/) Feb 28
Joe Hightower Mar 13
Tom Wentworth (http://www.cals.ncsu.edu/botany/Faculty/twentworth/twentworth.html) April 10
Rob Dunn (http://www4.ncsu.edu/%7Errdunn/) April 17
Homework 10
Critique 1 20
Critique 2 20
Proposal 30 Due April 30,2008
Exam 20 Due April 30,2008
Total 100
Homework Set 1-Due Tuesday January 29th.
Homework Set 2- Due Tuesday February 5.
Homework Set 3- Due Tuesday February 19
Due February 7, 2008
Structure Required-Each about 1-2 typed page double spaced for a total of 5-10 pages.
1. Summary of Key Points in the paper and critique
2. Scientific Significance of the Paper-Strengths and Weaknesses
3. Statistical Design Features-Strengths and Weaknesses
4. Statistical Analysis and Interpretation Features-Strengths and Weaknesses
5. References
Note- Some of you may choose "bad" papers to critique while others may choose very good ones. If possible try and pick papers that have interest to you scientifically as well as to satisfy this assignment
Paper Critique 2- Observational
Due March 18, 2008
Structure same as for first critique above
Proposals
Draft- April 1, 2008
Final- April 30,2008
Structure- I want you to write a proposal based on your real research proposal if at all possible. It will be like a standard proposal but with some more detail on design and analysis features. Please talk to me if you have any questions on the topic.
Later I will add some more information on the structure but it will be quite standard.
Guidelines
Exam
Take Home at end of semester. Handed out April15 and due April 30, 2008.
Audience:
Graduate students in ecological disciplines; also
applied
statistics and biomathematics
graduate students.Advanced undergraduates in
ecological
areas or statistics are also especially encouraged to take this class
and
can be admitted by special permission of the instructor.
Relationship to Other
Courses:
This course is different from other experimental
design
classes on campus in that it specifically focuses on
ecological applications and goes beyond the
traditional
experimental designs covered in ST 431 or ST 711.
Ecological field studies broadly divide into experiments (usually for small scale studies) and observational studies
(when the studies on larger spatial scales). We will begin with experimental concepts: use of controls, randomization, replication and blocking to improve precision. We will then consider situations where there are problems with designing true experiments in applied ecology.
In particular we will consider problems of lack of randomization and pseudo replication. We will also look at quasi-experimental designs and in particular BACI designs (Before After Control Impact Designs) for
assessing environmental impact.
We will also discuss the design of observational studies and exploratory methods (especially multivariate methods) for their analysis.
We will discuss important sampling design issues for obtaining ecological field data irrespective of whether we are using experiments or observational studies.
We will consider the special problems of meta analysis where multiple studies are combined into one overall analysis
All topics will be illustrated with reference to real ecological studies.
Overall References
(No Required Text)
Morrison, M. L. et
al.(2001). Wildlife
Study Design. Springer, New York
Scheiner et al. (2001). Design
and Analysis of Ecological Experiments. Oxford University Press. Second
Edition
Williams,
Nichols
and
Conroy
(2002). Analysis and Management of Animal Populations. Academic
Press.
ISBN 0-12-754406-2
Other General Design and Analysis Books
Manly,
B.F.J.(1992).The
Design and Analysis of Research Studies.Cambidge University Press.
Manly,
B.F.J.(2001). Statistics
for Environmental Science and Management. Chapman and Hall.
Ramsey and
Schaefer
(1997). The Statistical Sleuth. A Course in Methods of Data
Analysis.
Duxbury
Press
Steel et al. (1997).
Principles and Procedures of Statistics. 3rd Edn, McGraw Hill
Important Pseudoreplication Papers
Hurlbert (1984) Ecological
Monographs 54, 187-211.
Heffner et
al. (1996)
Ecology
77, 2558-2562.
Ecology special issue 1996 Volume 77 number 3.
There are several brief articles. Make sure you read Carpenter, Drake et al. and Lawton.
Important General References on Expts in Ecology
Lawton (1995) Science 269 328-331. Ecological expts with Model Systems.
Carpenter et al (1995) Science 269 324-327. Ecological expts with Model Systems.
Levin, S. (1992). The problem of Pattern and Scale in Ecology. Ecology 73, 1943-1967.
Important References on Quasi Experiments (BACI).
Chapter 6 of Manly (2001) referenced above.
Downs et al. (2002). Monitoring Ecological Impacts: Concepts and Practice in Flowing Waters. Cambridge University Press.
p 234-248.
Important General Sampling Books
Thompson, S. K. (2002) Sampling, 2nd Edition, Wiley, New York.
Thompson, S. K. and Seber, G. A.F.(1996) Adaptive Sampling, 2nd Edition, Wiley, New York.
Important References on Meta Analysis
1. Introduction
The Scientific Method, Theories, Hypotheses, Models
2. Statistical Definitions and Concepts
Populations and Samples
Point Estimation, Bias and Precision or Estimators
Confidence Interval Estimation
Hypothesis Testing, Type I and Type 2 Errors, Power of Tests
3. Classical Experimental Designs
Randomization, Replication and Blocking
Completely Random Design
Randomized Complete Block Design
Treatment Designs (Simple, Factorial)
Split Plot Designs
Repeated Measures Designs
4. Problems of Experimental Design in Ecological Field Studies
Lack of Randomization
Lack of Replication
Pseudoreplication
Spatial Scales of Interest vs Those That Are Practical
5. Quasi- Experimental Designs
Before After Control Impact Designs (BACI)
Weaker Designs
No Before Impact Designs
No Control Impact Designs
Other Designs
6. Design of Observational Studies
7. Sampling Design Issues for Experiments and Observational Studies
Simple Random Sampling Without Replacement
Systematic Random Sampling
Stratified Random Sampling
Cluster and Multi-State Sampling
Double Sampling
Adaptive Sampling
8. Exploratory Data Analysis in Observational Studies
Multiple Regression Analysis
Logistic Regression Analysis
Principal Components Analysis
Other Multivariate Methods (Brief)
9. Meta-Analysis (Combining Multiple Studies)
10. Integration of Field Studies, Mathematical Models, Adaptive Management