Department of Environmental Sciences
Huxley College of the Environment
Western Washington University

ESCI 340 Biostatistical Analysis
Winter 2016
MW 8-10 AM ES 80
F 8-10 AM AH 5

Instructor: John McLaughlin Teaching Assistant: Matt Sturza
Office: ES 434 Office: ES 303
Phone: 650-7617 Phone: 650-4416
E-mail:   E-mail: sturzam[at]students[dot]wwu[dot]edu
( Please do not send attachments in proprietary formats.)
Office Hours: MWF 10 Office Hours: MWF 11-12

Course Web Site: http://faculty.wwu.edu/jmcl/Biostat/syl2016w.htm
http://larch.huxley.wwu.edu/Biostat/syl2016w.htm
Please note: some hyperlinks may not be activated until class time.

Text: (Recommended) J. H. Zar. 2010. Biostatistical Analysis, 5thth ed. Prentice Hall
Additional readings available via links below.

Prerequisite: one year of general biology.

It is easy to lie with statistics. It is hard to tell the truth without statistics.
-- Andrejs Dunkels

Course Description:
This course is an introduction to data analysis and statistical tests commonly used in the biological and environmental sciences. Much of the material will be developed with a series of projects during which you will collect data to address research questions and analyze those data using appropriate methods. In a broad sense, the main objective of the course is to help you understand the principles, methods, and limitations of data analysis. After successfully completing the course, you should be able to identify appropriate applications of common statistical methods, to perform the methods competently, and to interpret statistical results critically.

Course Evaluation:
Grades will be based on homework assignments and three cumulative examinations. Homework assignments will comprise 50% of the course grade. The first two exams will contribute 15% each toward the course grade. The third exam will contribute the remaining 20%.

Homework:
Homework will be posted online to the links below, generally before class time on Friday.
Assignments are due at the start of class the following Friday. To earn credit, meet the deadline.

Homework Guidelines:
(1) Be clear, neat, complete, and concise. The teaching assistant has many assignments to grade;
it will be to your advantage to organize your work to make her job easier.

(2) Staple your work, if you submit more than one page.

(3) Put your name, course name, assignment number, and date submitted somewhere at the top of the first page.

(4) Show your work. Correct methods will be worth more than correct answers. For full credit, show all formulas used. Numerical tools (calculators, spreadsheets, SPSS, R) often combine several steps their calculations; you must show formulas for each step. When you use computer programs, indicate commands or menu options that you used to obtain your results.

(5) For assignments based on data collected for class, state assumptions that you made in your analysis.

Course Schedule: Please note: some hyperlinks may not be activated until after class.

Week Topics Research Project 
Jan. 6 Summary Statistics
Displaying data
Introduction to R
Computer lab transcript, Jan. 8
Edge effects on tree growth
Traffic loads on Bellingham/WWU streets (Snowy weather alternative)
Jan. 11 Distributions
estimation with uncertainty
Computer lab transcript, Jan. 15
Meet 8 am at Stair Sculpture, between AW, CF, ES
Maple seed dispersal distances
Jan. 18 Martin Luther King, Jr. Day -- No class
Jan. 20 Linear Models
Hypothesis testing: Comparing means, variances
Computer lab transcript, Jan. 22
Moss growth on maple trees
Avian scavenger abundance
Jan. 25 Hypothesis testing, continued:
Comparing means, variances
Exam 1: Wed. Jan. 27 (one page of notes permitted)
Exam 1 extra credit
study question answers
Computer lab transcript, Jan. 29
Feb. 1 Hypothesis testing: Proportions and frequencies
Computer lab transcript, Feb. 5
Maple seed dispersal distances
Feb. 8 Hypothesis testing: Regression and Correlation
Example: regression calculations w/ artificial "data"
Moss growth vs. tree size
Feb. 15 Presidents' Day -- No class
Feb. 17 Logistic Regression
Exam 2: Wed. Feb. 17 (two pages of notes permitted)
study question answers
More practice problems
Computer lab transcript, Feb. 19
Travel mode vs. distance
Stream channel stability vs. urban development
Feb. 22 Information Theoretic methods and Multimodel inference
Computer lab transcript, Feb. 26
The Mixed Nut Problem
Feb. 29 Course Review: Appropriate application of statistical methods; Answers
More review questions; Answers

Extra review sessions (optional)
Tue March 1, 9-10am, ES 410
Tue March 1, 5pm, ES 410
Thur March 3, 5pm, ES 410
Even more review questions

Exam 3: Friday March 4 (four pages of notes permitted)

***************************************************************************

Statistical Methods

Reading (from Zar)

Nomenclature
Distributions, random variables, and sampling
Selecting the appropriate hypothesis test

Ch. 1 - 6
Sections 24.1 - 24.3, 25.1

Critiques of statistical paradigms Stephens et al. 2006. Tr.Ecol.Evol. 22:192-197.
Lamblin 2012. Theory & Psychology 22:47-90
Anderson et al. 2000. JWM64:912-923.
Stephens et al. 2005. J.Appl.Ecol.42:4-12.
Nester 1996. Appl.Stat.45:401-410.
Introduction to R
R information and software downloads
USGS site about R
R homepage
Probability
Probability distributions in R
Further reading
One-sample hypothesis tests
One-sample t-test example, in R
Sections 7.1-7.5
Two- and paired-sample hypothesis tests
Two-sample t-test examples, in R
t-test examples using normal deviates, in R
Two-sample t-test spreadsheet example w/m&m data
Two-sample t-test example w/m&m data, in R
Paired-sample t-test example, in R
Variance test example, in R
Sections 8.1-8.3, 8.5; 9.1, 9.2
Parametric vs. nonparametric tests
Mann-Whitney Two-sample test example, in R
Wilcoxon paired-sample test example, in R
Sections 8.9, 8.10, 9.5
Statistical Power Sections 7.6, 8.4, 9.3

Analysis of Variance
Single-factor ANOVA
single-factor ANOVA example, in R
Multiple comparison test
Tukey multiple comparison example, in R
Tukey multiple comparison calculator
Nonparametric ANOVA; Bartlett's test (variances)
Bartlett's test calculations, in R
Two-factor ANOVA


Sections 10.1 - 10.3, 10.6
Sections 11.1 - 11.3
Section 10.4
Sections 12.1-12.3, 12.6, 12.7
Goodness of fit
Chi-squared goodness of fit test example, in R
Kolmogorov-Smirnov goodness of fit test example, in R
Contingency tests
Sections 22.1 - 22.5, 22.7 - 22.9
Sections 23.1 - 23.7
Simple Linear Regression and Correlation
Simple linear regression
Regression worksheet
Simple linear correlation
Data transformations
Data transformation example, in R

Ch. 17, Sections 18.1 - 18.3
Sections 19.1 - 19.6
Ch. 13
Logistic regression example, in R
Evaluating Multiple Hypotheses
Akaike's Information Criterion
Multimodel Inference
Model selection example
Model selection example, in R
Model selection: tidepool example, in R
Model selection: mixed nut example, in R
Multimodel inference, variable importance: tidepool example, in R
Anderson et al. 2000. JWM64:912-923.
A&B 2002. JWM66:912-918.
Burnham & Anderson 2002. ch.8
Likelihood Methods
Further reading:
Hobbs NT, Hilborn R. 2006. Alternatives to statistical hypothesis testing in ecology: a guide to self teaching. Ecological Applications 16(1):5-19.
Readings TBA

***************************************************************************

Homework Assignments
Assignments are due at 0800 on the date listed.

Assignment

Due Date

One Jan. 15
Two Jan. 22
Prepare for exam 1
Three Feb. 5
Four Feb. 12
Prepare for exam 2
Five Feb. 26
Six March 4

Return to Courses