Module DXX-2022:
Quantitative Thinking

Module Facts

Run by School of Natural Sciences

20 Credits or 10 ECTS Credits

Semester 1 & 2

Organiser: Dr Isabel Rosa

Overall aims and purpose

This module aims to provide students with the necessary tools to confidently deal with basic numeracy, different data types and their manipulation, and importantly, how to convert data commonly used in environmental and ecological sciences into meaningful information and how to communicate it clearly and concisely. Building on the quantitative methods learnt in in the 1st year, the students will consolidate their practical skills, particularly targeting environmental and ecological data analysis, to plan and perform a scientific analysis and communicate the results in a clear format. Using a series of practical case studies addressing real-world data analyses, the students will critically evaluate multiple data types (categorical and continuous) and utilise quantitative tools to creatively stimulate their quantitative thinking (problem-solving skills) on how to convert these data into relevant information. Using a practical approach, students will learn a set of transferable skills such as the ability to conduct critical analyses covering a range of applications relevant to environmental sciences, from land use change, to animal behaviour tracking, to species richness patterns, to socio-economic surveying, and forest inventory data, to name a few. The core of the module is in ensuring that the students are able to critically look at different datasets typically used in environmental and ecological sciences (from a practical and academic perspective) and translate these into meaningful information that can be shared with multiple audiences.

Course content

Real-world case studies of environmental data analysis will be used to demonstrate the need to confidently translate unstructured data into relevant information. In particular, the case studies will cover a wide range of topics to give the students a broad view on what kind of problems and techniques they might be encounter throughout their careers in environment-related jobs. In detail, the generic contents of this module are:

  • Introduction to environmental and ecological data science (what do the different data types mean?): understanding typical data structures, for instance species occurrences matrices, animal tracking data, remote-sensing data, forest inventory data, surveys. Importantly, how can we formulate and answer to research questions based on these data structures.
  • Introduction to programming in R, as a tool to be used throughout the whole module. Why R? Open-source, reproducible science.
  • Pre-analysis: common data processing techniques (manipulating data, organizing tables, etc.) when dealing with environmental and ecological datasets. Ensuring a good data organisation prior to the start of any statistical analysis.
  • Summarising data: understanding the need to comprehend and describe the datasets (via descriptive statistics such as central tendency and spread, i.e, mean, median, variance), prior to perform statistical analysis.
  • Data visualization techniques: translating unstructured data into quality figures (graphical methods for data exploration), using the right plot to answer the right questions, e.g. boxplots, histograms, scatterplots, line and bar plots.
  • Statistical analyses: correlation (and why it does not mean causation); generalized linear models and statistical tests (parametric and non-parametric, including data transformation). Finding the strength/confidence on the answers to the posed research questions.
  • Science communication tools: how to write a report and how to do an effective presentation, addressing both an academic (e.g., scientific report) and a non-academic audience (e.g., policy brief). How to clearly convey the message of your environmental data analysis.

These contents will be explored within the context of 'real-world' examples, such as within the context of environmental impact assessments (before-and-after analyses), analysing species richness and species trends over time, monitoring timber production, etc. Invited speakers will give the students a wide perspective on the importance of confidently dealing with data. Apart from academic staff, other invited speakers include non-academics working for organisations such as the British Trust for Ornithology, Euroforest, Natural Resources Wales.

Assessment Criteria

good

Grade C- to B+ Able to organise and summarise the data properly. Appropriate choice of data representation. Good understanding of how to choose the most appropriate statistical test for the analysis to be performed. Good scientific report, well structured and with clearly described methodology, and description of main findings. Good ability to translate scientific work into non-expert communication. Evidence of ability to perform a complete data analysis task from the initial pre-processing of the raw data, to appropriate methodological procedures, to communication of main findings. Good domain of the R language.

excellent

Grade A- to A**. Excellent understanding of the research questions to pose, and ability to process and analyse the data to answer such questions. Excellent organisation of scientific report, with clear descriptions of objectives, methodology adopted and results. Strong domain of the R language, with well written scripts, at times even going beyond the suggested solutions. Evidence of independent data exploration and analysis with own suggestions of potential methodologies to employ. Clear translation of scientific work to non-expert audiences. Demonstration of an exceptional ability in producing well-structured and accurate data representations (plots) for the presentation of results from various analyses. Very high standard of presentation.

threshold

Grade D- to D+ Adequate knowledge of data organisation and description. Low ability to choose the correct data representation and/or the appropriate statistical test to answer research question. Report containing a poorly described methodology and/or errors. Low quality presentation both in the scientific report and in the non-scientific course work (policy brief or whitepaper). Poor ability to use R to perform the coursework laid out in class.

Learning outcomes

  1. Be able to summarise and describe the dataset (presentation)

  2. To confidently process and organise a dataset prior to any analysis

  3. To learn how to use R confidently

  4. Be able to carry out a analysis and communicate it clearly in the form of a scientific report

  5. Be able to translate a scientific report into a policy brief or a non-expert document (whitepaper)

  6. Be able to select the appropriate graphical representation of the data

Assessment Methods

Type Name Description Weight
FORMATIVE ASSESSMENT Understanding how to work with categorical data (analysing surveys)

The students will be asked to return an R script showing how they imported, processed, plotted and performed basic statistical analyses on categorical data. In detail, the students will analyse timber growth in young, mid-, and old-growth plots.

20
CASE STUDY Understanding and working with continuous data

The students will be asked to return an R script where they imported, processed and used basic statistical analysis to correlate two independent continuous variables, as well as produce a trend over time using linear regression. In particular, the students will analyse species trends over time, i.e. choose a species and analyse whether the poulation has been growing or decreasing over time in the UK.

20
CASE STUDY Understanding the basics of R programming and data analysis

In this assessment, the students will be asked to write a short R script to important and process a data table, produced a couple of plots and describe the data and plots using summary statistics. The students will also be asked to write one sentence that describes the plot produced.

20
CASE STUDY Data science: download, process, analyse and communicate

The students will be asked to conduct a full (short) data science 'project' on the last couple of practical sessions of the module. The will collect data (download from freely available online data), process it, plot it, and conduct the statistical analysis. The topics will be restricted to those similar to what was introduced to the students throughout the module, and the research questions will be discussed in class prior to the coursework to ensure manageable projects to be done within the time allocated. The students will be then be asked to give a short presentation (selecting either an academic or non-academic audience) presenting the results and write a very short report on what they have done.

40

Teaching and Learning Strategy

Hours
Practical classes and workshops

Practical case studies (some with invited guest lecturers) showing how to perform a particular data analysis (within the context of conservation science). For instance, using remote sensing-driven data to understand land use change dynamics, or tracking animal behavior, or working with forest inventory data, or assessing species richness in different transects, etc.

36
Lecture

Introduction of the concepts to perform data analysis

12
Private study

Work on assigned coursework, i.e. R scripts to perform certain data organisation, representation (plots) as well as an individual scientific project and presentation to non-expert audience

152

Transferable skills

  • Literacy - Proficiency in reading and writing through a variety of media
  • Numeracy - Proficiency in using numbers at appropriate levels of accuracy
  • Computer Literacy - Proficiency in using a varied range of computer software
  • Exploring - Able to investigate, research and consider alternatives
  • Information retrieval - Able to access different and multiple sources of information
  • Critical analysis & Problem Solving - Able to deconstruct and analyse problems or complex situations. To find solutions to problems through analyses and exploration of all possibilities using appropriate methods, rescources and creativity.
  • Presentation - Able to clearly present information and explanations to an audience. Through the written or oral mode of communication accurately and concisely.

Subject specific skills

  • PS1 CommunicationSkills,Covering both written and oralCommunication with a variety of audiences
  • PS2 Skills in the employment ofCommonConventions andStandards inScientific writing, dataPresentation, and referencing literature
  • PS3 Problem-solvingSkills, relating to qualitative and quantitative information
  • PS4 numeracy and mathematicalSkills, including handling data, algebra, functions, trigonometry,Calculus, vectors andComplex numbers, alongside error analysis, order-of-magnitude estimations,Systematic use ofScientific units and different types of dataPresentation
  • PS5 information location and retrievalSkills, in relation toPrimary andSecondary informationSources, and the ability to assess the quality of information accessed
  • PS6 information technologySkills whichSupport the location, management,Processing, analysis andPresentation ofScientific information
  • PS11 Problem-solvingSkills including the demonstration ofSelf-direction, initiative and originality
  • PS15 the ability to thinkCritically in theContext of data analysis and experimental design
  • SK20. Development of generalStrategies including the identification of additional information required andProblems where there is not a uniqueSolution.

Resources

Reading list

The R book: https://www.wiley.com/en-gb/The+R+Book%2C+2nd+Edition-p-9780470973929;

The Elements of Statistical Learning: https://web.stanford.edu/~hastie/ElemStatLearn/

Remote Sensing and GIS for Ecologists: http://book.ecosens.org/

R graphics cookbook: http://www.cookbook-r.com/Graphs/

R for Data Science: https://r4ds.had.co.nz/

Pre- and Co-requisite Modules

Courses including this module

Optional in courses: