Course Co-ordinator: Dr David Lusseau; Dr Rene Van der Wal
Pre-requisite(s): An undergraduate statistics course.
Note(s): BI5509 is the code for students starting their programme in January.
The module will be divided in themed weeks during which students will gain skills in sampling design (through practicals) and analytical technique (through lecture and computer labs).
Week 1: introduction to biostatistics. Students are introduced to simple sampling design, core statistical concepts, and statistical software.
Week 2: introduction to statistical modelling. Students continue their progression in statistical analyses and are introduced to complex sampling design.
Week 3: generalised linear models. Students learn about generalised linear models and the interpretation of models (model fitting, model selection, and model validation) and are exposed to more advanced models. Students carry out sampling in groups for their report.
Week 4: categorical data. Students learn about statistical techniques for categorical data. They also learn about power analyses to understand the influence of sample size on tests results.
Week 5: multivariate statistics. Students cover multivariate statistical techniques and are given the opportunity to go over material covered in previous weeks.
Week 6: student-lead teaching. Students are given the opportunity to go over previous material to reinforce learning and are given time to prepare their report.
Three 3-hour lectures per week. One 8-hour practical session each week including field trip and computer lab.
The module will be assessed based on 2 graded practicals (20% each) and an independent report (60%)
Course Co-ordinator: Drs David Lusseau, Rene van der Wal and Alex Douglas
Pre-requisite(s): BI5009 or equivalent
The module will be divided into themed weeks during which students will gain skills in sampling design (through practicals) and analytical techniques (through lecture and computer labs).
Week 1: introduction to complex study design - students are introduced to nested and repeated sampling and random effects.
Week 2: dealing with complex design in linear models - students learn to account for complex sampling and effects in linear models using linear mixed effect models, generalised least squares models and generalised additive models.
Week 3: correlated data structure - students are introduced to spatial and temporal autocorrelation in lectures and in practicals and learn ways to deal with it in linear models.
Three 2-hour flexible lecture/tutorial slots each week (thread II)
One eight-hour practical session each week including field trip and computer lab (thread II)
The module will be assessed based on an independent report (100%).
Course Co-ordinator: Drs Thomas Cornulier, Alex Douglas, David Lusseau
Week 1: Introduction to Bayesian statistics. After a refresher in probability theory and linear modelling, students are introduced to Bayes theorem, Bayesian inference, and estimation tools.
Week 2-3: Bayesian implementaion of models for various study designs. Students will learn to implement statistical models in the R/BUGS language and fit them to ecological data. Students will gain experience in the visualisation and validation of models and focus on their ecological interpretation. Students will start by implementing relatively simple models that they have already covered using a frequentist approach in previous statistics courses (BI5009 and BI5010) and progress to models suited to more advanced study designs.
Five 3-hour lecture sessions combined with practical applications and research seminars (thread 2).
1 online assessment via myaberdeen and 1 marked practical.
Course Co-ordinator: Dr Rene van der Wal
An introduction to the course and the reading list;
independant study with support from mentor;
oral presentations and oral examinations
evaluation of the exercise.
The course will run over 3 weeks in thread 2; in the first week there will be a two hour introductory session, followed by a two hour discussion of topics later in the week. In the second week there will be individual meetings between students and staff. In the third week there will be a half day session for oral presentations and then individual oral exams with the course coordinator and other members of the teaching team.
100% continuous assessment (divided as follow: oral presentation (50%), oral examination (50%).
Course Co-ordinator: Dr Jo Smith
Aims: To provide the student with the skills to understand and learn how to use techniques for Ecological and Environmental modelling, including model design and evaluation.
Learning Objectives: By the end of this course you should be able to:
1. understand the criteria for design or choice of models, and the use of models in relation to selected policy issues;
2. use Excel for model development, use a number of dedicated models, and apply statistical methods to evaluate models;
3. choose appropriate models for a given application; develop a model and use it to critically assess and ecological/environmental issue; critically assess limitations reliability and applicability of models; and, formulate scientific concepts in a mathematical form.
A total of 7 one-hour lectures, 2 three-hour computer classes and 2 one-hour tutorials.
Continuous assessment by 2 written problem-solving assignments and 1 essay.
Course Co-ordinator: Dr Michelle Pinard
How to plan a project; how to write a research proposal
There will be one 3-hour session in each of the three weeks where the class meets with the co-ordinator. The other contact time will be arranged with project supervisors within the School. These contact hours with staff occur informally currently; this course would only serve to formalise the arrangements and provide a structured assessment to assist the students to progress with their project planning.
100% continuous assessment in the form of a proposal