Transcript Lecture 1 - introduction to course
Data Handling & Analysis BD7054 2012-2013
Andrew Jackson Zoology, School of Natural Sciences [email protected]
Statistics in science
• • • Analysis of data is central to science – Metaphorically – Literally (Introduction -> methods -> results -> discussion) Underpins one’s own research – Your own research project Essential in understanding others’ research – To question what they did – To incorporate their ideas in your own research
The scientific method
• Ask a question about the world around you – Why are vultures the only obligate scavengers among the extant terrestrial vertebrates?
The scientific method
• • • • Decide what measurable outcome you will use to test a specific hypothesis – Physiology of vultures favours this mode of life – Compare metabolic costs of flight across vertebrate taxa Design an experiment or field study to test this idea Use statistics to determine whether your predictions hold Frame your findings within the broader background of the precedent science –introduction and discussion
Course Outline
• • • 8 th Oct – 12 th Oct – Introduction to R and statistics 21 st Jan – 25 th Jan – General Linear Models 29 th April – 3 rd May – Generalised Linear Models – Multivariate methods
Assessment
• On the Friday ending each week, you will be asked to submit either – an assessment, or – complete an online exam assessing your proficiency in data analysis using R
Learning outcomes
• • • • •
NB slightly different from course handbooks
summarise and communicate quantitative results graphically and textually to scientific standards.
apply appropriate statistical analyses of commonly encountered data types.
discuss the context of the analyses within a hypothesis driven framework of scientific logic.
use the R statistical computing language for data analysis.
Course structure
• • • Series of Lecture / tutorials and computer practicals Lectures will be as interactive as possible Computer practicals – Use R to analyse data – Follow video podcasts for instruction – Demonstrators present to help
Week 1
• • Lectures / Tutorials – Monday to Thursday 10-12 – GGSR-A Computer sessions – Monday to Thursday 14-16 – Botany Hut computer rooms
Summary of statistics covered
• • • Linear regression General linear models – As a way to ask increasingly complex questions of our data using a common framework (ANCOVA / multiple regression) Generalised linear models – Extending these concepts to deal with non-normal data types (binary / surivival / count data)
Statistical software - R
• • • R is a command line interfaced software – – Scary the first few times Incredibly powerful and adaptable – – Free Open development Time-tabled computer sessions – Complete video-podcast and examples in your own time When Googling for R related topics add “cran” to your search terms
Delivery of course content
• • • Attendance at lectures/tutorials is mandatory Moodle website associated with course – Lectures will be posted – Web-based discussions – Links to video-podcasts Statistics, An introduction using R. Michael J Crawley. Wiley. ISBN 0-470-02298-1
Basic Experimental Design
For more details see Experimental Design for the Life Sciences by Ruxton and Colegrave
Relationship between hormone levels in male chimpanzees and #females • • • Measure hormone levels of male chimps and then count how many females are they foraging with.
Higher hormone levels are expected when there are more females to mate with.
However, hormone levels are influenced by age, diet, time of day etc.
Male hormones and #females
• Hormone level difference could be due to age, diet, time of day OR #females
Relationship between hormone levels in male chimpanzees and #females • All chimps are the same age, diet, and time of day so hormone level difference ~ #females
Class Exercise
Come up with a scientific question and plot your predictions
Computer Session 8
th
October
• Work through 3 podcasts on my website – http://www.tcd.ie/Zoology/research/research/the oretical/Rpodcasts.php
1. Opening R for the first time 2. Working with script files 3. Importing data into R