BIO 621: Current Topics in Biology

Biometry: The Design and Analysis of Biological Experiments
Syllabus - Spring Semester 2012

Meeting Time: TR, 11:00am-12:15pm, Room 205, T H Morgan Bldg

Instructor: Craig Sargent

Prerequisites: STA 570 or equivalent, or consent of the instructor. STA 671/672 (or equivalent) is highly recommended.

Textbooks: This class builds on the content of STA 671/672 , where we explore Mixed Models and Generalized Linear Models, which both use Maximum Likelihood methods rather than Least Squares methods. We will use the following book:

Quinn and Keough (2002), Experimental Design and Data Analysis for Biologists, Cambridge (link) Author Websites: Quinn, Keough

Software: We will be using mostly JMP, and some SAS, to solve problems in class. JMP doesn't do Generalized Linear Mixed Models, and for that application, we will use SAS's Proc GLIMMIX.

SSTARS at the University of Kentucky administers site licenses for SAS, JMP and SPSS (and some other software), and you can obtain annual individual licenses through them. R is open source and free, and is extremely powerful if you know what you're doing. 

SAS  Online Procedures Documents: 

Learning Objectives: After this class, students should feel comfortable reading the statistical procedures from research publications in their field; designing, analyzing and presenting their own research experiments; and, delving more deeply into the applied statistical literature as necessary.

Course Description: This class is taught in an inverted format. Students will work through PowerPoint lectures as homework, and do statistics problems together on laptops in class. This class emphasizes application and presentation of the statistical procedures that are commonly encountered in biology. Students will be presented the full experience of designing experiments, analyzing data (provided by the instructor or the student), and presenting the statistical results in standard scientific format (tabular, graphical). Students will work with datasets from their own research specialties. We will review General Linear Models, and ordinary least squares approaches to Analysis of Variance and variations on that theme: regression, ANOVA, ANCOVA, and MANOVA. Then we will focus on linear models that use maximum likelihood parameter estimation, such as mixed models, and generalized linear models, including logistic models. Students are welcome to explore more advanced topics for their projects.

Grading: There will be regular homework assignments (ungraded), two graded take-home exams (worth 30% per exam), and a final project (worth 40%).

Data Sets: Excel Files: Face Widths..,

Class Notes: pdf files of the class PowerPoint slides

Exam 1: Dataset: Key:

Exam 2: Dataset: Key

Projects:

Below is an approximate schedule (which is a work in progress).