Syllabus

Description

Statistics are a key component of rigorous science and as such there is a need to both understand advanced statistics and properly document the analysis to improve scientific communication transparency and quality. The course aims to:

  1. Provide an understanding of advanced statistical models (including generalized linear mixed models)
  2. Develop good coding practices (using R and Rmarkdown)
  3. Improve data and code management (data manipulation and github)
  4. Present the principles of open science (using OSF, Open Science Framework).

Contents

  • Statistics (flexible to be define based on attendance interests)
    • Mixed models including generalized linear mixed models and multivariate models
    • Options:
      • Introduction to Bayesian stats
      • Path analysis
      • Spatial analysis
      • Multivariate analysis
      • …
  • Open and reproducible science
    • R: Rmarkdown, tidyverse and workflowr
    • git: github and other alternatives
    • Open Science Framework with OSF

Classes

Lecturer

  • Julien Martin

Office hours

Meetings will be in-person or online using zoom. Check brightspace for the links

  • Julien Martin: Wednesday 1100-1200 (in-person & online)

Schedule

  • Lecture: Tuesdays, 1430-1720 in CRX-C323

There are no lectures or tutorials during reading week.

See the course website for details of the schedule

Assessment

Based on interactions in classes and on a data analysis project using R and documented with with Rmarkdown, workflowr and github.

  • 50% in class assessment:
    • 15 min presentation in small groups on a paper and questions (30%)
    • discussion in class (10% questions prep, 10% discussions)
  • 50% written statistical report using analysis more complex than linear models written using Rmarkdown, published on github in 2 different formats

Presentations

Register team, article and date

Please register for your presentations here.

You need to log into your uottawa account to have access

General info

The presentation counts for 30% of the final grade

The presentation should be based on one of the suggested paper or paper selected by the students (with my approval)

Presentation should be between 8-10 minutes long and are followed by 3-5 minutes of questions. They should provide a big picture of the study and the main hypothesis, a quick presentation of the data, detailed explanation of analysis and results, and finally a discussion on the clarity and adequacy of the statistical analysis. If the paper is more methodological, the presentation should present a summary version of the paper and discuss what what harder to understand or not understood.

Rubric

Content – 60 marks

  • Introduction: Background of the general ‘big picture’ and specific problem, study system, and specific hypotheses or predictions.
  • Methods: Brief outline of data collection and explanation of statistical methods used to address the hypotheses/predictions outlined in the introduction, including key assumptions.
  • Results: Visualizations clearly summarizing your results, using parallel structure to methods
  • Discussion: Summary of major results, alternative explanations for the observed results, conclusions/implications, future directions

Style – 30 marks

  • Organization
  • Clarity
  • Delivery: Timing, pace, tone, articulation
  • Originality: Creativity, critical thinking

Answers and understanding – 10 marks

Written Report

General info

The written report counts for 50% of the final grade

Using your own dataset (or one provided by me):

  1. develop a question of interest
  2. choose the statistical analysis appropriate to answer the question
  3. analyse the data
  4. write a short report using Quarto/Rmarkdown explaining in details:
    • the analysis done
    • the results
    • model assumptions and associated checks
    • the limitation of the model or data to answer the question
  5. submit your work via Brightspace before teh deadline

You need to submit a reproducible document. Your submission on Brightspace needs to include:

  1. rendered pdf (could be a rendered docx saved as pdf)
  2. .[Rq]md file
  3. data file(s)
  4. extra files needed to reproduce the document
    • .bib files
    • .csl files
    • …

Missing files would incur a 10% penalty on the report grade

Rubric

Content – 70 marks

Introduction (10 marks)
  • Background of the general ‘big picture’ and specific problem,
  • study system
  • specific hypotheses to be tested
  • predictions
Methods (20 marks)
  • Brief outline of data collection
  • clear explanation of statistical methods used to address the hypotheses/predictions outlined in the introduction, including key assumptions.
Results (25 marks)

Text and visualizations clearly summarizing your results, using parallel structure to methods. Should include :

  • clear graphics representing the data and the model outputs,
  • tables reporting statistics
  • text describing results
  • evaluation of model assumption and model fit
Discussion (15 marks)
  • Summary of major results
  • conclusions/implications
  • Limitations of analysis (power, complexity of model, data structure, …)

Style – 30 marks

Clarity and reproducibility of code (10 marks)
  • code well written and understandable
  • document written in Quarto/Rmarkdown, reproducible with .[qR]md and data
  • only necessary code included
  • all necessary code included
Organization and formatting (10 marks)
  • clear and logical
  • correct formatting of the document
Clarity of text (10 marks)
  • clear and understandable
Typos and grammar (5 marks)
  • it is all in the section title

Marking disputes

If you feel an error has been made in the grading of an assignment or exam, you have 10 working days from the date the item was returned to request a regrading. Beyond 10 working days the mark is considered final. If the issue concerns a lab assignment, please consult the TA that marked it; if it concerns and exam, please see the course instructor.

When a regrading occurs, the new mark will stand whether it is higher or lower, so there is no guarantee your mark will increase. Simple adding mistakes unfortunately happen on occasion given the volume of assignments and exams we have to mark and such cases are easy to correct. However, disputes over partial marks on partially correct answers will not always work out in your favour and should be reserved for situations in which you feel you have a strong case. When requesting a regrading, you must provide an explanation as to what you feel the issue is (i.e. why you think you deserve a higher grade). The more specific this is the better your chance of success.

I do not consider arguments for additional marks simply because you are close to the next letter grade.

Plagiarism

Plagiarism is the act of passing off someone else’s words or ideas as your own. Plagiarism is a major academic offense (see the University of Ottawa regulation on academic fraud: Academic fraud) and is also illegal. We uphold the law. It is critical that you understand what constitutes plagiarism and how you can avoid it. For more information, consult the following document from the Student Academic Success Service.

Attendance

As per uOttawa policy, to ensure they succeed in all courses of their program of study, students have the responsibility to participate in the various learning and assessment activities for this course. Attendance of the lectures and laboratory are not formally required for this course, although both are STRONGLY recommended. The supplied lecture slides will not include all content presented and discussed in class, and teaching assistants and the professor are available during the labs for help with exercises and assignments. TA and professor office hours are to allow you an opportunity to ask questions related to course material and cannot replace a missed lecture or lab.

Bilingualism

Please note that the class is bilingual. This means that any questions during the class can be asked or answered in either french or english and the prof will translate the answers if needs be. The majority of the course content (slides, lab manual) are provided in english only, but explanation can be provided in either languages.

In addition, as per uOttawa policy, all students have the right to produce their written work and to answer examination questions in the official language of their choice, regardless of the course’s language of instruction.