Statistics logo
Department of Statistics & Actuarial Science
Simon Fraser University
Author: Carl James Schwarz  
Professional Statistician
P.Stat. (Statistical Society of Canada), PStat® (American Statistical Association)
Phone: Retired.
Office: Retired.
Email: Email

Course Notes for Intermediate Ecological Statistics

The course notes below illustrate methods of analysis using JMP, R, or SAS. Instructions for installing R, JMP, SAS (and other softeware)..

Sample code used in the notes is available at: Sample Program Library. Note that even though a chapter may not have a version for a package, the program code for the example is often available -- I just haven't had time yet to update the notes to include the code and output directly in the notes.

The suggested citation for a chapter of notes is:
      Schwarz, C. J. (2014). Chapter Name.
      In Course Notes for Beginning and Intermediate Statistics.
      Available at http://www.stat.sfu.ca/~cschwarz/CourseNotes. Retrieved yyyy-mm-dd.

     Package Chapter and sections
     JMP R SAS 1 In the beginning...
                1.1 My favorite papers
                1.2 Introduction
                1.3 Effective note taking strategies
                1.4 It's all $\Gamma \rho \epsilon \epsilon \kappa $ to me
                1.5 Which computer package?
                1.6 FAQ - Frequently Asked Question
     JMP R SAS 2 Introduction to Statistics
                2.1 TRRGET - An overview of statistical inference
                2.2 Parameters, Statistics, Standard Deviations, and Standard Errors
                2.3 Confidence Intervals
                2.4 Hypothesis testing
                2.5 Meta-data
                2.6 Bias, Precision, Accuracy
                2.7 Types of missing data
                2.8 Transformations
                2.9 Standard deviations and standard errors revisited
                2.10 Other tidbits
     JMP R SAS 3 Sampling
                3.1 Introduction
                3.2 Overview of Sampling Methods
                3.3 Notation
                3.4 Simple Random Sampling Without Replacement (SRSWOR)
                3.5 Sample size determination for a simple random sample
                3.6 Systematic sampling
                3.7 Stratified simple random sampling
                3.8 Ratio estimation in SRS - improving precision with auxiliary information
                3.9 Additional ways to improve precision
                3.10 Cluster sampling
                3.11 Multi-stage sampling - a generalization of cluster sampling
                3.12 Analytical surveys - almost experimental design
                3.13 References
                3.14 Frequently Asked Questions (FAQ)
     JMP R SAS 4 Designed Experiments - Terminology and Introduction
                4.1 Terminology and Introduction
                4.2 Applying some General Principles of Experimental Design
                4.3 Some Case Studies
                4.4 Key Points in Design of Experiments
                4.5 A Road Map to What is Ahead
     JMP R SAS 5 Single Factor - Completely Randomized Designs (a.k.a. One-way design)
                5.1 Introduction
                5.2 Randomization
                5.3 Assumptions - the overlooked aspect of experimental design
                5.4 Two-sample $t$-test- Introduction
                5.5 Example - comparing mean heights of children - two-sample $t$-test
                5.6 Example - Fat content and mean tumor weights - two-sample $t$-test
                5.7 Example - Growth hormone and mean final weight of cattle - two-sample $t$-test
                5.8 Power and sample size
                5.9 ANOVA approach - Introduction
                5.10 Example - Comparing phosphorus content - single-factor CRD ANOVA
                5.11 Example - Comparing battery lifetimes - single-factor CRD ANOVA
                5.12 Example - Cuckoo eggs - single-factor CRD ANOVA
                5.13 Multiple comparisons following ANOVA
                5.14 Prospective Power and sample sizen - single-factor CRD ANOVA
                5.15 Pseudo-replication and sub-sampling
                5.16 Frequently Asked Questions (FAQ)
                5.17 Table: Sample size determination for a two sample $t$-test
                5.18 Table: Sample size determination for a single factor, fixed effects, CRD
                5.19 Scientific papers illustrating the methods of this chapter
     JMP R SAS 6 Single factor - pairing and blocking
                6.1 Introduction
                6.2 Randomization protocol
                6.3 Assumptions
                6.4 Comparing two means in a paired design - the Paired $t$-test
                6.5 Example - effect of stream slope upon fish abundance
                6.6 Example - Quality check on two laboratories
                6.7 Example - Comparing two varieties of barley
                6.8 Example - Comparing prep of mosaic virus
                6.9 Example - Comparing turbidity at two sites
                6.10 Power and sample size determination
                6.11 Single Factor - Randomized Complete Block (RCB) Design
                6.12 Example - Comparing effects of salinity in soil
                6.13 Example - Comparing different herbicides
                6.14 Example - Comparing turbidity at several sites
                6.15 Power and Sample Size in RCBs
                6.16 Example - BPK: Blood pressure at presyncope
                6.17 Final notes
                6.18 Frequently Asked Questions (FAQ)
     JMP R SAS 7 Incomplete block designs
                7.1 Introduction
                7.2 Example: Investigate differences in water quality
     JMP R SAS 8 Estimating an overall mean with subsampling
                8.1 Average flagellum length
     JMP R SAS 9 Single Factor - Sub-sampling and pseudo-replication
                9.1 Introduction
                9.2 Example - Fat levels in fish - balanced data in a CRD
                9.3 Example - fat levels in fish - unbalanced data in a CRD
                9.4 Example - Effect of UV radiation - balanced data in RCB
                9.5 Example - Monitoring Fry Levels - unbalanced data with sampling over time
                9.6 Example - comparing mean flagella lengths
                9.7 Final Notes
     JMP R SAS 10 Two Factor Designs - Single-sized Experimental units - CR and RCB designs
                10.1 Introduction
                10.2 Example - Effect of photo-period and temperature on gonadosomatic index - CRD
                10.3 Example - Effect of sex and species upon chemical uptake - CRD
                10.4 Power and sample size for two-factor CRD
                10.5 Unbalanced data - Introduction
                10.6 Example - Stream residence time - Unbalanced data in a CRD
                10.7 Example - Energy consumption in pocket mice - Unbalanced data in a CRD
                10.8 Example: Use-Dependent Inactivation in Sodium Channel Beta Subunit Mutation - BPK
                10.9 Blocking in two-factor CRD designs
                10.10 FAQ
     JMP R SAS 11 Two-factor split-plot designs
                11.1 Introduction
                11.2 The three basic structures
                11.3 Data and labeling experimental units.
                11.4 Assumptions
                11.5 Example - Tensile strength of paper - main plots in CRD
                11.6 Example - Biomass of trees - main plots in an RCB
                11.7 Example - Tenderness of meat - main plots in an RCB
                11.8 Example - Fungi degrading organic solvents - a split-plot in time
                11.9 Example - Home range - an unbalanced split-site plot in time
                11.10 Example - Floral scents and learning - pseudo-replication
                11.11 Example - Pheromone effects upon wild type and anarchist colonies of bee
                11.12 Repeated Measure Designs analyzed as a Split-Plot Analysis
                11.13 Example - Holding your breath at different water temperatures - BPK
                11.14 Example - Systolic blood pressure before presyncope - BPK
                11.15 Final notes
                11.16 Frequently Asked Questions (FAQ)
     JMP R SAS 12 Analysis of BACI experiments
                12.1 Introduction
                12.2 Before-After Experiments - prelude to BACI designs
                12.3 Simple BACI - One year before/after; one site impact; one site control
                12.4 Example: Change in density in crabs near a power plant - one year before/after; one site impact; one
                12.5 Simple BACI design - limitations
                12.6 BACI with Multiple sites; One year before/after
                12.7 Example: Density of crabs - BACI with Multiple sites; One year before/after
                12.8 BACI with Multiple sites; Multiple years before/after
                12.9 Example: Counting fish - Multiple years before/after; One site impact; one site control
                12.10 Example: Counting chironomids - Paired BACI - Multiple-years B/A; One Site I/C
                12.11 Example: Fry monitoring - BACI with Multiple sites; Multiple years before/after
                12.12 A statistical diversion
                12.13 Closing remarks about the analysis of BACI designs
                12.14 BACI designs power analysis and sample size determination
     JMP R SAS 13 Comparing proportions - Chi-square ($\chi ^2$) tests
                13.1 Introduction
                13.2 Response variables vs. Frequency Variables
                13.3 Overview
                13.4 Single sample surveys - comparing to a known standard
                13.5 Comparing sets of proportions - single factor CRD designs
                13.6 Pseudo-replication - Combining tables
                13.7 Simpson's Paradox - Combining tables
                13.8 More complex designs
                13.9 Final notes
                13.10 Appendix - how the test statistic is computed
                13.11 Fisher's Exact Test
     JMP R SAS 14 Correlation and simple linear regression
                14.1 Introduction
                14.2 Graphical displays
                14.3 Correlation
                14.4 Single-variable regression
                14.5 A no-intercept model: Fulton's Condition Factor $K$
                14.6 Frequent Asked Questions - FAQ
     JMP R SAS 15 Detecting trends over time
                15.1 Introduction
                15.2 Simple Linear Regression
                15.3 Transformations
                15.4 Pseudo-replication
                15.5 Introduction
                15.6 Power/Sample Size
                15.7 Power/sample size examples
                15.8 Testing for common trend - ANCOVA
                15.9 Example: Degradation of dioxin - multiple locations
                15.10 Example: Change in yearly average temperature with regime shifts
                15.11 Dealing with Autocorrelation
                15.12 Dealing with seasonality
                15.13 Seasonality and Autocorrelation
                15.14 Non-parametric detection of trend
                15.15 Summary
                15.16 ?
     JMP R SAS 16 Regression with pseudo-replication
                16.1 Introduction
                16.2 Example: Selenium concentration in fish tissue
                16.3 Pseudo-replication when regression is over time
                16.4 Comparing slopes after environmental impact
     JMP R SAS 17 Regression - hockey sticks, broken sticks, piecewise, change points
                17.1 Hockey-stick, piecewise, or broken-stick regression
                17.2 Searching for the change point
                17.3 What is the first time that a treatment mean differ from a control mean
     JMP R SAS 18 Analysis of Covariance - ANCOVA
                18.1 Introduction
                18.2 Assumptions
                18.3 Comparing individual regression lines
                18.4 Comparing means after covariate adjustments
                18.5 Power and sample size
                18.6 Example: Degradation of dioxin - multiple locations
                18.7 Example: Change in yearly average temperature with regime shifts
                18.8 Example - More refined analysis of stream-slope example
                18.9 Example: Comparing Fulton's Condition Factor $K$ among groups
                18.10 Final Notes
     JMP NA NA 19 Multiple linear regression
                19.1 Introduction
                19.2 Example: Blood pressure vs.\ age, weight, and stress
                19.3 Regression problems and diagnostics
                19.4 Polynomial, product, and interaction terms
                19.5 The general linear test
                19.6 Indicator variables
                19.7 Example: Predicting PM10 levels
                19.8 Variable selection methods
     JMP R SAS 20 Regression - hockey sticks, broken sticks, piecewise, change points
                20.1 Hockey-stick, piecewise, or broken-stick regression
                20.2 Searching for the change point
                20.3 What is the first time that a treatment mean differ from a control mean
     JMP R SAS 21 Logistic Regression
                21.1 Introduction
                21.2 Data Structures
                21.3 Assumptions made in logistic regression
                21.4 Example: Space Shuttle - Single continuous predictor
                21.5 Example: Predicting Sex from physical measurements - Multiple continuous predictors
                21.6 Retrospect and Prospective odds-ratio
                21.7 Example: Parental and student usage of recreational drugs - $2 \times 2$ table.
                21.8 Example: Effect of selenium on tadpoles deformities - $2 \times k$ table.
                21.9 Example: Pet fish survival - Multiple categorical predictors
                21.10 Example: Horseshoe crabs - Continuous and categorical predictors.
                21.11 Assessing goodness of fit
                21.12 Variable selection methods
                21.13 Complete Separation in Logistic Regression
                21.14 Final Words
     JMP R SAS 22 Logistic Regression - Advanced Topics
                22.1 Introduction
                22.2 Sacrificial pseudo-replication
                22.3 Example: Fox-proofing mice colonies - dealing with sacrificial pseudo replication
                22.4 Example: Over-dispersed Seed Germination Data
                22.5 Example: Are mosquitos choosy? A preference experiment.
                22.6 Example: Reprise: Are mosquitos choosy? A preference experiment with complete blocks.
                22.7 Example: Reprise: Are mosquitos choosy? A preference experiment with INCOMPLETE blocks.
     JMP NA NA 23 Poisson Regression
                23.1 Introduction
                23.2 Experimental design
                23.3 Data structure
                23.4 Single continuous $X$ variable
                23.5 Single continuous $X$ variable - dealing with overdispersion
                23.6 Single Continuous $X$ variable with an OFFSET
                23.7 ANCOVA models
                23.8 Categorical $X$ variables - a designed experiment
                23.9 Log-linear models for multi-dimensional contingency tables
                23.10 Variable selection methods
                23.11 Summary
     JMP R SAS 24 A short primer on residual plots
                24.1 Linear Regression
                24.2 ANOVA residual plots
                24.3 Logistic Regression residual plots - Part I
                24.4 Logistic Regression residual plots - Part II
                24.5 Poisson Regression residual plots - Part I
                24.6 Poisson Regression residual plots - Part II
     JMP NA NA 25 Time Series - a VERY brief introduction
                25.1 Introduction
                25.2 Fundamental material
                25.3 White noise and autocorrelation in regression
                25.4 Detrending, Differencing and Integration
                25.5 Autoregressive Models on stationary series
                25.6 Moving Average Models on stationary series
                25.7 Combining Moving Average and Autoregressive Models - \ensuremath
                25.8 Model Selection - I
                25.9 Estimation
                25.10 Model Selection - II - AIC
                25.11 Model checking
                25.12 Forecasting
                25.13 Summary
     JMP R SAS 26 Tables
                26.1 A table of uniform random digits
                26.2 Selected
                26.3 Selected
                26.4 Cumulative probability for the
                26.5 Selected percentiles from the
                26.6 Selected percentiles from the
                26.7 Sample size determination for a two sample $t$-test
                26.8 Power determination for a two sample $t$-test
                26.9 Sample size determination for a single factor, fixed effects, CRD
                26.10 Power determination for a single factor, fixed effects, CRD
     JMP R SAS 27 THE END!
                27.1 Statisfaction - with apologies to Jagger/Richards
                27.2 ANOVA Man with apologies to Lennon/McCartney
     JMP R SAS 28 An overview of environmental field studies
                28.1 Introduction
                28.2 Analytical surveys
                28.3 Impact Studies
                28.4 Conclusion
                28.5 References
                28.6 Selected journal articles
                28.7 Examples of studies for discussion - good exam questions!

Reanalysis of data from published papers

Material not yet integrated into the above notes

Short courses that are available - send me an email for details.

These are typically 3 days in length and can be offered as a block of 3 days, or spread over several weeks (e.g. 1/2 day per week over 6 weeks). Many of these have been offered via the
Columbia Mountains Institute of Applied Ecology. I can also give these short courses at your own site. Unfortunately, SFU's Distance Education faculty has shown NO interest in offering these through SFU.

Ministry of Forestry Publications

Additional Teaching materials - past exams; sample mid-terms; assignments, etc. These materials are graduatlly being merged into my general course notes

  • StatVillage - a clickable map using real census data
  • On-line experimental design - a system for generating computer experiments
    [ SFU Statistics and Actuarial Science Home | SFU Home]
    Email comments or suggestions to Carl Schwarz (cschwarz@stat.sfu.ca)
    © 2012 Carl James Schwarz Last updated 2012-01-04