Course Coordinators
Prof. Moshe Kiflawi, The Interuniversity Institute for Marine Sciences in Eilat .
Prof. Roi Holzman, Faculty of Life Sciences and the Steinhardt Museum of Natural History Tel Aviv University and the Interuniversity Institute for Marine Sciences Eilat.
Lecturers
Prof. Moshe Kiflawi, The Interuniversity Institute for Marine Sciences in Eilat .
Prof. Roi Holzman, Faculty of Life Sciences and the Steinhardt Museum of Natural History Tel Aviv University and the Interuniversity Institute for Marine Sciences Eilat.
Course Brief Description
The course consists of two preliminary meetings and an intensive week at the IUI.
The preliminary meetings (about 5 hours each) will be held in Zoom and will include lectures and computerized practice. These meetings will take place on Sunday March 9 and Sunday March 16, 2025. The participation in the preliminary meetings is mandatory.
The week in Eilat will include lectures, computer exercises, fieldwork and data-analyses. The days are typically long, often extending into the night. Throughout the week the students work in small groups and present their results at the end of each section.
Aims
To get the most biological insight out of your hard-earned data, you need good insight into the statistical analyses with which you will analyze them. Following this rationale, the course will expand on the basics laid in introductory Biostatistics courses, and introduce some of the more advanced analytical tools which are the staple of quantitative ecological research.
Major topics covered in the course:
A re-examination of key concepts in statistics: p-values, parameters and their estimates, sampling distributions, hypothesis-testing, effect-size, randomization tests.
Issues in experimental design; namely: Replication, sample size & power-analyses.
How to validate and curate a data-set.
Assumptions of the general-linear model and what to do when the data deviate from them:
- Normality & heterogeneity of variance: generalized-linear-models for binary & count data
- Hierarchical structure in the data: mixed-effects models
How to deal with multiple predictor variables:
- Multiple regression & co-linearity.
- Model-selection (AIC, model-averaging)
Reading:
Zuur, A.F., Ieno, E.N. & Elphick, C.S. (2010) A protocol for data exploration to avoid common statistical problems. Methods in Ecology & Evolution, 1, 3– 14.
Pre-requisites
The course is an advanced course that requires a strong quantitative approach.
- Participants are required for a broad background in ecology that includes at least one advanced course in ecology.
- An advanced course in statistics that provides comprehensive knowledge of statistical techniques of T-tests, chi square tests, variance analysis (anova) and regression is also required.
- The course requires a basic acquaintance with the R environment - for example: an academic R course or the following online course https://www.codecademy.com/learn/learn-r (the online course is given for free by registering for a trial period on the website and at the end of it you can get a certificate of completion of the course). Students with demonstrable programming skills (in R) may be exempt, following consultation with the course coordinators.
- Reading: Zuur, A.F., Ieno, E.N. & Elphick, C.S. (2010) A protocol for data exploration to avoid common statistical problems. Methods in Ecology & Evolution, 1, 3– 14.
Grading
10% Participation
10% Preparatory exercise, given during the preliminary meeting
80% Final report.(Unusually requires extensive work). The report is submitted a month after the course ends.
Target audience
The course is limited to 18 M.Sc. and Ph.D. students.
Language
English. Reports can be submitted in English or Hebrew.