BSc Biology

Analysing, interpreting and communicating quantitative data is a central skill for biologists. Data literacy is not only the bread and butter of understanding scientific evidence in published research. In an age of “big data” and social media (mis/dis)information, a critical eye for what the data can and can’t tell us is key. 

In this course you will use data exploration for generating hypotheses and learn the basics of data visualization. You will learn how researchers use data to tell a “story” of discovery and practice articulating such a story in a structured way. Key concepts of experimental design, sample distribution and hypothesis testing will be revisited and practiced, along with graph creation with suitable software, including R. 

To build competency in identifying types of variables; understanding samples, experimental design and data structure. To understand how exploratory analysis of large data sets can generate hypotheses. To practice and build confidence in and choosing appropriate graphs and statistical tests using suitable software, including R. To appreciate and apply the principles of good data visualization. To develop competency in articulating concisely and accurately what quantitative data show.

Students will be able to:
• Explore larger data sets in HHMI Data explorer through visualizations and statistical tests; analyse and interpret the data and generate hypotheses
• Choose appropriate statistical tests for hypotheses derived from various types of experimental or observational data (including revisiting key statistical concepts from Year 1)
• Use suitable software, including R, for conducting statistical tests, where necessary using cleaned, transformed and/ or processed data, and for graph production
• Communicate descriptive statistics and test results accurately and concisely, and articulate effectively what a data figure shows
• Produce professional graphs, applying basic principles of good data visualization (key concepts data-ink ratio, chart junk, lie factor, data density)  
• Work with authentic experimental data from FBMH research and complete processing, analysis, visualization and evaluation in the form of a short result section
• Work with selected open-source ‘big data’, collaborate on data wrangling, graph production and ‘data storytelling’ 
 

Online material is designed to revisit key statistical concepts and provide guided exercises in producing graphs with R (“Graph of the week”). Three opportunities for troubleshooting the installation and use of R are scheduled.  
 

Face-to-face sessions are a mixture of interactive lectures and Team-based learning  workshops:
• Data exploration: Exploring and visualising multivariate data with HHMI data explorer; understanding variables; generating and testing hypotheses. 
• Workshops: Understanding stats concept, recognizing data structure, communicating results, building hypotheses, graphing and stats.
• Introduction to research data workshop (bioscience background).
• Workshops: analysing authentic research data and communicating the findings.
• Principles of good data visualization.
• Workshops: Introduction to open data (ecological, public health etc); wrangling, exploring and visualising data; telling a story with data.
 

Online stats quizzes: Formative
“Graph of the week” submissions: 10%
Minireport from research data OR R-based ‘data story’ and commentary: 40%
Stats and experimental design exam: 50%
 

Formative feedback is provided in stats e-learning modules and “graph of the week” exercises, as well as during TBL sessions. A skills audit (in-class MCQ) at the start gives students a concrete idea of where their weak spots are. Instructors in workshop provide ad hoc formative feedback on group work in progress.
 

Feedback on the summative assignments (report or ‘data story’) will be given on Blackboard. 
 

tbc