SCIU7SR - Statistics Using R

CO-ORDINATOR: Dr Luc Bussière
CONTRIBUTORS: Dr Nils BunnefeldDr Timothy Paine


This module will introduce advanced methods in linear and nonlinear model building, adopting examples from recent biological and environmental science literature. We will use the open-access R software platform to ensure that these skills are portable regardless of the eventual career choice of our students.

Module Objectives

  • To review the philosophical issues concerning the assessment of systematic and random error in empirical science
  • To emphasize the distinction between and relative value of parameter estimation and null hypothesis testing in modern statistical analyses
  • To help develop skills in the graphical representation of hypotheses, and appreciate how to confront these graphical models with empirical observations.
  • To clarify how one may translate predictions of covariance and difference into predictions concerning parameters of linear (or linearized) equations.
  • To clarify modern approaches to model simplification and evaluation
  • To practice critical assessments of statistical procedures in published and unpublished work
  • To clarify best practice in the presentation of analyses for professional audiences

Learning Outcomes

On successful completion of this course, students will;

  • Be competent in the basic use of R (a statistical analysis package) and R studio (a client that makes coding in R easier) for statistical analysis of biological or environmental data, and be familiar with the potential practical applications of the software.
  • Understand how to undertake effective manipulation, statistical analysis, interpretation and presentation of data, and how to critically assess the same procedures in the scientific literature

Key Generic Skills:

  • IT
  • Oral communication
  • Basic Statistical Analysis
  • Written communication
  • Data Quality awareness
  • Time management
  • Critical thinking
  • Creative presentation


The learning of key software packages and their effective application necessitates a significant practical component to this course.  The course is taught through lectures that will introduce the key theoretical aspects of data analysis and manipulation.  The module also features a heavy emphasis on a weekly pair of computer-based practical laboratory sessions of up to three-hours duration each.  The second of these consist of help clinics that will typically include time for reviewing material covered during the first session, but they may occasionally include new material.  It is strongly advised that you attend all practical sessions and lectures, no matter what your background is.

Course Assessment

Assessment will be based on an open-book computer-based practical class test delivered around the midsemester mark (40%), and a final project that includes both a quantitative review of statistical practice in a subdiscipline chosen by the student (typically this will be the subdiscipline most closely aligned with the student’s honours work; 20%) and analyses of new data collected by the student in that subdiscipline (40%). The standard expected for this work will be in line with the full-module status of the course.

© University of Stirling FK9 4LA Scotland UK • Telephone +44 1786 473171 • Scottish Charity No SC011159
Portal Logon