Syllabus Biometry for Coastal Sciences Fall 2018

**Class 01 – Introduction to data and statistical analysis**

- Reading: Broman and Woo (2018) Data Organization in Spreadsheets
- The nature of statistical analysis
- Objectives of statistical analysis
- Description and inference
- Objects, variables, and scales
- Types of biometric data

- Variable classes

Lecture 01 Biometry Lecture 01 Why do we need statistics

**Class 02 – Frequency Distributions and Variation**

- Frequency and cumulative frequency
- Theoretical frequency distributions
- Normal Distribution

- Measures of central tendency
- Variation and associated measures

**Class 03 – Probability, Sampling, and Parameters**

- Probability
- Random variables
- Sampling from a normal distribution
- Sampling from a bionomial distribution
- Parameter Estimation
- Student’s
*t*-distribution

**Class 04 – In-class assignment**

For your outside of class review and consideration:

**Class 05 – Hypothesis Testing and Statistical Power**

- Reading:
- Trafimow et al. 2017 Trafimow 2017

Benjamin et al. 2018 BenjaminEtAlRedefineStatisticalSignificance

- Trafimow et al. 2017 Trafimow 2017
- The null hypothesis
- Tests of the mean
- Significance level
- Type 1 and 2 error
- Difference between means and proportions
- Power
- Assumptions of parametric statistics

Link to zipped file with the R code for: power analysis function

**Class 06 – In-class assignment**

**Class 07 – Correlation**

- Patterns of correlation
- Indicators
- Interpretation and calculation of the correlation coefficient,
*r* - Rank correlation

Clarification on determining the critical value of the t distribution for null hypothesis significance testing (best resource I have found):

http://janda.org/c10/Lectures/topic06/L24-significanceR.htm

**Class 08 – Midterm 01**

Exam 01 Answer Key from 2017

**HW01: **

**Kachigan Exercises (p. 495):****Note if you don’t have time or desire to do all of the these then only do those questions that have answers to ‘selected’ exercises (Kachigan 546, Zar “Answers to selected exercises”). I give you this large HW assignment because I believe it is the best way to learn the material. Only the selected exercises will be graded.**- Ch. 1: 1, 2, 3, 4, 5, 6, 7, 8 (use internet), 12
- Ch. 2: 1, 3 (use < 15 words), 4, 5, 6, 7, 8-20
- Ch. 3: 1, 2, 3, 5, 6, 7, 8 (use technical terminology), 9, 10, 11 to 13, 15, 16 (but explain using your statistical knowledge and investigation, why pie charts should be avoided)
- Ch 4: 1 to 25
- Ch 5: 1 to 1 to 14, 17, 18, 19, 20, 23, 26, 27, 28 (this is z-score!), 29 (and this is one reason we use them), 29 to 32
- Ch 6: 1 to 3, 5 to 7, 26, 27
- Ch 7: 1 to 6, 12 to 15, 17
- Ch 8: 1 to 7
- Ch 9: 1, 2, 3, 4, 5, 7, 8, 9, 10, 11, 12, 13 to 16, 19, 20, 21, 22, 23, 24, 30, 31, 32, 33, 34, 35, 36
- Ch 10: 1-15

- Zar (4th edition)
- 6.2, 6.3
- 7.2, 7.4

**Reflective reading**- 250 words focusing on Broman and Woo (2018)
- 500 words compare and contrast Trafimow et al. 2017 and Benjamin et al. 2018
- What is reflective reading? See this document: Reflective reading

**Class 09 – Simple Linear Regression**

- Regression model
- Fitting
- Confidence intervals and description of variance
- Residuals
- Predictors and Scaling

Lecture 09 Simple Linear Regression (Excel Worksheet)

**Class 10 – Multiple Linear Regression**

- Regression model
- Fitting
- Dummy variables
- Multi-colinearity

**Class 11 – In class assignment**

Correlation practice

Correlation and Power In class assignment

correlation and power (access this file to get the function for the simulation)

Differences in power among statistical tests

**Class 12- In class assignment**

Simple and Multiple linear Regression

In_class_assignment_12 (Simple Linear Regression)

In_class_assignment_12b (Multiple Linear Regression)

**Class 13 – Comparing two mean values**

*t*-test

Biometry Lecture 13 Comparing two means

**Class 14 – In-class assignment**

In-Class Assignment 14 Comparisons of two means

**Class 15 – One-way ANOVA**

**Biometry Lecture 15 Comparing multiple means one-way ANOVA**

**Class 16 – In-class assignment**

**Class 17 – Factorial ANOVA**

- Statistical modeling
- Biometry Lecture 17 Factorial ANOVA

*Also:*

*ANOVA as a linear model video – I recorded this for a previous class and thought it may be informative to you. Let me know if you found it helpful!*

Here is the text (slightly modified) used in the “*ANOVA as a linear model*” video

**Class 18 – Midterm 02**

Exam 02 Answer Key from 2017

Exam 02 Answer Key from 2018

**HW02:**

**Note if you don’t have time or desire to do all of the these then only do those questions that have answers to ‘selected’ exercises (Kachigan 546, Zar “Answers to selected exercises”).**- Fields
- Ch. 9 Task 1, Task 2 (End of chapter)
- Ch. 10 Task 1, Task 3 (End of chapter)

- Kachigan
- Ch. 11 (All Selected Exercises)

- Zar (4th edition)
- 17.1
- 17.3
- 20.1 (must show that data can be used for parametric analysis)
- 20.3
- 20.4

**Class 19 – Analysis of Categorical Data**

Biometry Lecture GOF and Contingency Tables

**Class 20- In class assignment**

**Class 21- Non-parametric statistical methods**

Biometry Lecture 20 Non Parametric Tests

**Class 22- In class assignment**

To be posted

**Class 23- Computer-intensive methods**

In_class_assignment_(show and tell)

In_class_assignment_(show and tell)

**Class 24- Eating Turkey**

**Class 25- Non-linear curve fitting and AIC**

Leaf and Murphy Case study data

**Class 26- Midterm 03**

HW 1 of 3: Focus 100% of your commentary on the statistical evaluation of these works. Take the role as a reviewer of the statistical approaches on these methods. Really looking forward to seeing your thoughts – ps. Increasing sample size is rarely if ever the answer. No minimum page/word number – just give me your best effort, max would be about 750 words or so total. Choose one: 1 and 2, or 3 below.

- https://rgs-ibg.onlinelibrary.wiley.com/doi/full/10.1002/geo2.44
- https://rgs-ibg.onlinelibrary.wiley.com/doi/full/10.1002/geo2.48
- Morley et al 1983

HW 2 of 3:

Zar (4th edition) 22.1, 22.2, 22.3, 22.4

HW 3 of 3:

- Use non-linear curve-fitting techniques to answer question #4 (at the end of the case study).
- Fit at least three different models to these data.
- Use AIC to evaluate alternative model fit. If the delta AIC value among competing models is < 4, use a model weighing approach to determine predicted weight-at-length.

**Class 27- Review**

**Class 28- Final Exam**