Use the Tab and Up, Down arrow keys to select menu items.
Advanced experimental design and statistical techniques for biologists. This course is essential for all students considering postgraduate study in biological sciences.
This course provides a detailed understanding of how the interpretation and analysis of data depends on the way the data were collected. It covers various aspects of experimental design, a variety of statistical methods to analyse data, basic programming in R, and places a heavy emphasis on interpreting the results of analysis. It is intended for anyone who would like to know more about designing robust research, analysing datasets of various kinds, or who wishes to better identify the misleading use of statistics in the media or in research. The experimental design component is particularly suited for anyone who plans to do postgraduate research in any theme of biological sciences, or who would like to pursue a career in any field where it is useful to identify patterns in data.
As a student in the course, I will develop the ability to:1. Identify potential analysis approaches, and apply a wide range of statistical tests (including linear models, non-parametric tests, and generalised linear models) to answer scientific questions (assessment tasks: Lab quizzes, Structured learning assessments, Concept tests). (Graduate attribute: Employability, innovation and enterprise).2. View statistical methods as an inter-related set of tools that can be applied to different situations, and to be able to defend your choice of analysis for a given problem (assessment tasks: Lab quizzes, Structured learning assessments, Concept tests). (Graduate attribute: Employability, innovation and enterprise).3. Design robust experiments and research using observational data, and interpret the results of analysis in light of experimental design (assessment tasks: Lab quizzes, Structured learning assessments, Concept tests). (Graduate attribute: Employability, innovation and enterprise);4. Carry out analyses and basic programming in the R statistical package (assessment task: Lab quizzes). (Graduate attribute: Employability, innovation and enterprise)5. Interpret data with an understanding of how certain we can be about what we know, and how hidden variables may cloud our judgement or give a false impression of causation (assessment tasks: Lab quizzes, Structured learning assessments, Concept tests). (Graduate attribute: Employability, innovation and enterprise).6. Leverage the strengths of AI, while understanding its limitations, having the skills to critique its outputs, and understanding the ethical and scientific issues surrounding its use in data analysis. (assessment tasks: Lab quizzes, Concept tests). (Graduate attributes: Globally aware, Biculturally confident).Pūkenga Ngaio | Transferable Skills As a student in this course, I will develop the following skills:• The ability to phrase statistically rigorous, biologically interesting hypotheses.• The ability to identify the best experimental design to test specific questions.• Proficiency with a diverse array of statistical tests and data manipulations in the R programming environment.• The ability to interpret statistical results presented in scientific papers.• The ability to communicate the biological meaning of statistical tests and accurately describe methods of data analysis.• The ability to integrate AI into my workflow while quality controlling the results.(Graduate attribute for all transferable skills: Employability, innovation and enterprise)
BIOL209 or appropriate statistical background as determined by the Head of School
Students must attend one activity from each section.
Summary of the Course ContentSome topics covered by this course are:o Very brief revision of ANOVA, Regressiono Multifactor ANOVA, Multiple regressionso General linear models and dummy variableso Assumptions of parametric tests and dealing with data that don’t meet themo Designing biological studies (experimental design, nested and split plot designs)o Generalised linear models using Poisson and Binomial errorso Analysis workflow involving AI, reproducibility, data provenance and sovereignty.o Multivariate statistics (Clustering, Ordination, and PERMANOVA)For more information on lecture topics see the Course Outline below.
Jason Tylianakis
Helen Warburton
This course will not directly follow any text book, and the lab manual has detailed explanations for much of the material. However, if you’d like a book to help with your study or for reference:Crawley, M.J. (2015) Statistics: An Introduction using R. (2nd Ed.) Wiley & Sons, Chichester.is available from the University Bookshop and UC library, is user-friendly and gives a good overview of the material covered in the course. You should be aware, however, that no book is good at explaining everything that might be helpful to a statistically-minded biologist; additional references may be recommended and put on restricted loan in the library or on Learn.Other potential referencesQuinn, G. P. & Keough, M. J. (2002). Experimental Design and Data Analysis for Biologists. Cambridge University Press, Cambridge.Additional resources will be made available during the course on Learn.
Learn Site Course Outline
Feedback from previous Course Survey 2021Student ratings (out of 5) The materials provided helped me to understand what was required to succeed in this course. 4.51 The organisation of this course helped me learn. 4.49I found the workload was appropriate to the level of the course. 4.36I found the assessments appropriate for the course. 4.36Where I sought feedback on my assessments, I found it helpful. 4.49Helpful features1. Lecture slides were informative and easy to follow, along with a good explanation of each slide.2. The 309 lab manual has been extremely useful in understanding the course content.3. Very good organisation of the course, small frequent quizzes and labs with recap sessions at the end helped a lot!4. The lecture content order seemed good.5. A lot less stressful than the assessments for 209, I really appreciated the types of assessments being changed to smaller but more frequent assignments. Helped me learn more as I was less stressed.6. The online lab tests were good, they sound a lot better than a big lab exam, and they helped me learn the lecture material throughout the term rather than waiting for exam time.7. The nature of the internal tests seemed fair as it would've been very difficult to memorise all the R coding, especially as it comes later with continual use.What to change? (Action/response indicated in capitals)The lectures where well organised but the lecture slides online had important parts of it deliberately left blank. This makes it a bit harder to go back through notes and reinforce learning.THIS IS DELIBERATE, BECAUSE RESEARCH SHOWS THAT IF YOU MAKE NOTES AND WRITE THINGS DOWN (RATHER THAN JUST LISTENING OR READING), YOU RETAIN THE INFORMATION BETTER. HOWEVER, WRITING TOO MUCH DOWN DURING A LECTURE CAN PREVENT YOU FROM HAVING ANY TIME TO STOP AND THINK. AS A COMPROMISE, WE GIVE MUCH OF THE LONG TEXT AND FIGURES IN THE POWERPOINTS, BUT LEAVE BLANK THE KEY WORDS FOR YOU TO WRITE THEM DOWN YOURSELF TO HELP YOU REMEMBER THEM.But the labs we worth so few % and had tasks that were either extremely simple or extremely difficult no in between. I think it would perhaps be better to have a series of labs over the course to learn information as you go, or two weekly quizzes worth 5% or maybe even less just because they really help you learn as you go.WE ALREADY HAVE APPROXIMATELY TWO-WEEKLY TESTS WORTH 5%, SO NOT SURE HOW TO ADDRESS THIS. LAB TESTS HAVE SOME EASY QUESTIONS SO THAT HOPEFULLY NOBODY GETS ZERO, AND THEN PROGRESSIVELY MORE DIFFICULT QUESTIONS. THE R CODE TO ANSWER THEM ALWAYS COMES FROM A PREVIOUS LAB.
Domestic fee $1,099.00
International fee $5,388.00
* All fees are inclusive of NZ GST or any equivalent overseas tax, and do not include any programme level discount or additional course-related expenses.
For further information see School of Biological Sciences .