Main content

Course outline


Loading wiki pages...

Wiki Version:
**`IMPORTANT NOTE: The dates and times of the course are for 2022, but the content in the course outline has not yet been updated`** ## **Synthesis Statistics in Ecology & Evolution** #### **Instructors:** - Dr. Laura Pollock (McGill University) - laura[dot]pollock[at]mcgill[dot]ca - Dr. Jennifer Sunday (McGill University) - jennifer[dot]sunday[at]mcgill[dot]ca - Postdocs from the Living Data Project: Gracielle Higino, Mike Lavender, Sam Straus #### **Course Description** Synthesis statistics for ecology and evolution This course will provide an introduction and overview of approaches for synthesizing the highly structured, multi-sourced datasets that typify ecology, evolution, and environmental research, including data collation, integration, analysis, and visualization. Students will learn how to define their research question and scope in the context of other studies in their subfield and how to find and fill in gaps in understanding with additional datasets. We will also provide a broad overview of the different types of methods used for synthesizing data including hierarchical models, meta-analysis, model integration, and model updating, providing students a guide to navigating these methods and identifying methods to learn and use in their research. Students will develop skills in R programming, collaborative research, reproducible workflows, data analysis, and communication. #### **Meeting Times:** Tuesdays and Thursdays - 8:00 am - 9:30 am PDT (British Columbia) - 9:00 am - 10:30 am CST (Saskatchewan) - 11:00 am - 12:30 pm EDT (Ontario/Quebec) **Week 1**: Tues Oct. 04, Thurs Oct. 06 **Week 2**: Tues Oct. 11, Thurs Oct. 13 **Week 3**: Tues Oct. 18, Thurs Oct 20 **No classes**: Tues Oct. 25, Thurs Oct. 27 **Week 4**: Tues Nov. 01, Thurs Nov. 03 #### **Delivery format** - 8 sessions, 1.5 hours per session - each session generally includes a lecture and hands-on component, each varying in length among sessions #### **Pre-requisites** - Graduate-level thesis in Ecology or Evolutionary Biology - Introductory coding experience or a willingness to learn #### **Required materials** - Personal computer (laptop or desktop) - internet connection - Please download and install [R]( and [RStudio]( before the first class #### **Online Resources:** All materials for this and associated courses will be available on course Canvas websites, supplemented by wiki pages the [OSF site]( for the Living Data Project. #### **Assessment** - [Assignment 1]( : 20% - [Assignment 2]( 70% - Participation: 10% #### **Workload** In general, students in each Living Data Project module should anticipate approximately 30-45 work hours, in line with the normal expectations for one credit of coursework. This will comprise the following activities: 12 hours formal instruction in a class setting, 4 hours of individual/small group mentoring, 5 hours of preparatory reading, up to 15 hours on required assignments. Assignments are completed progressively as the course proceeds to allow instructor feedback on drafts, with a final version of major projects due within a week of course completion Participation in this course requires adherence to the Living Data Project [Code of Conduct]( *** ### Tentative schedule of topics and activities (8 sessions) * * * `This schedule was last updated on 30 November 2020` * * * #### **Session 1: Tuesday, September 29th** **Suggested readings:** **Note for students**: all pre-readings (and papers used throughout the course are accessible from "Session materials - private" folder in [Class Communication]( - Hampton, S.E., Anderson, S.S., Bagby, S.C., Gries, C., Han, X., Hart, E.M., Jones, M.B., *et al*. (2015). The Tao of open science for ecology. *Ecosphere*. [PDF]( - Stewart, G. (2010). Meta-analysis in applied ecology. *Biology Letters*. [PDF]( **In-session:** 1. **Introductions** (10 min) - Meet the teaching team! - **Instructors**: Laura Pollock, Jennifer Sunday - **Postdocs**: Sarah Amundrud, Joey Burant, Ellen Bledsoe - **Teaching assistants**: Bruno Carturand, Mauro Sugawara 2. **Lecture: Jennifer Sunday** (20-25 min) - Overview of the course and expectations - Introduction to synthesis statistics - See: [session 1 lecture slides]( 3. **Activities** (45-50 min) - Overview and example (5 min) - Break-out groups: conceptualizing your research (15-20 min) - Link to activity will be provided in Zoom chat - Access the [completed activity]( - Class discussion (20 min) - Time permitting (if you haven't already done so): - Ensure [R]( and [RStudio]( are installed - Set up [OSF account](, including adding affiliation info - See [Instructions for students]( for requesting access to the Class Communication component and setting up your Submissions folder. - Postdocs and TAs are available for help! * * * #### **Session 2: Thursday, October 1st** **Pre-session readings:** **Note for students**: all pre-readings (and papers used throughout the course are accessible from "Session materials - private" folder in [Class Communication]( - Law, Y.-H. (2018). Replication failures highlight biases in ecology and evolution science. *The Scientist*. [PDF]( - Sharpe, D., and Poets, S. (2020). Meta-analysis as a response to the replication crisis. *Canadian Psychology*. [PDF]( **In-session:** 1. **Q & A / Admin** (5-10 min) - Address questions 2. **Lecture: Laura Pollock** (25-30 min) - Replicability, heterogeneity, and data synthesis - See: [session 2 lecture slides]( 3. **Discussion** (45 min) - Based on pre-readings - How might synthesis address the replication crisis in ecology and evolution? - See: [replication crisis discussion questions]( **Suggested follow-up reading:** - Ortega, Z., Martín-Vallejo, J., Mencía, A., Purificación Galindo-Villardón, M., and Pérez-Mellado, V. (2015). Introducing meta-partition, a useful methodology to explore factors that influence ecological effect sizes. *PLoS ONE*. [PDF]( * * * #### **Session 3: Tuesday, October 6th** **Pre-session readings:** **Note for students**: all pre-readings (and papers used throughout the course are accessible from "Session materials - private" folder in [Class Communication]( - Gurevitch, J., Koricheva, J., Nakagawa, S., and Stewart, G. (2018). Meta-analysis and the science of research synthesis. *Nature*. [PDF]( - Nakagawa, S., Noble, D.W.A., Senior, A.M., and Lagisz, M. (2017). Meta-evaluation of meta-analysis: ten appraisal questions for biologists. *BMC Biology*. [PDF]( **Further reading:** M. Borenstein, L.V. Hedges, J.P.T. Higgins and H.R. Rothstein (2009) A basic introduction to fixed-effect and random-effects models for meta-analysis. *Research Synthesis Methods*. [PDF]( **In-session:** 1. **Q & A / Admin** (5-10 min) - Address questions 2. **Lecture: Jennifer Sunday** (25-30 min) - Accounting for issues of publication bias, noise, and non-independence - understanding how sources of variation can be accounted for in meta-analysis and aggregated data analysis - See: [session 3 lecture slides]( 3. **Activities** (45 min) - Break-out groups: extracting relevant summary information for a meta-analysis (30 min) - Class discussion (15 min) - See: [meta-analysis summary table activity](; papers needed for the activity are available [here]( **Assignment** (post-session): - **Option 1**: Take the first steps of a meta-analysis by building a table that summarizes 3-5 relevant studies in your field. For this option, the emphasis will be on the table, but in text, provide the context and motivation (i.e. introduction) for the research question that this meta-analysis would answer, discuss the consistency or variation in the study statistics reported, and the modifiers that you hypothesize to be important. - **Option 2**: Find examples of (i) a meta-analysis paper and (ii) a primary data aggregation paper in your field of interest. Identify their questions, datasets, and reported statistics, and summarize these in a table or series of comparative figures. Consider how the two approaches were able to ask different kinds of questions. - **Due date**: please upload your assignment to your personal OSF Submissions component **by 10pm ET on Tuesday** (Oct. 13th). (McGill students should submit using [MyCourses]( - **Format**: We expect the assignment documents to be short, maximum 2 pages double spaced, normal-sized margins, 12pt font. Submit as a .pdf to maintain page formatting. Use a consistent reference style, title your report, and state the option number (1 or 2) and your name in the upper margin. - **Rubric**: We've created a [rubric]( for each option. * * * #### **Session 4: Thursday, October 8th** **Pre-session readings:** - readings today **In-session:** 1. **Q & A / Admin** (5-10 min) - Address questions 2. **Lecture: Laura Pollock and Jennifer Sunday** (30 min) - Review of common types of statistical terms - Introduction to effect sizes - What to extract from studies into tables - See: [session 4 lecture slides]( 3. **Activities** (45 min) - data wrangling using `tidyr` and `dplyr` - Cleaning data in a script - Merging datasets - See: tidy+dplyr [introduction]( and [tutorial]( **Suggested follow-up material to tutorial** - [Starting with data]( - [Manipulating, analyzing, and exporting data with tidyverse]( - [10 up-to-date ways to do common data tasks in R]( * * * #### **Session 5: Tuesday, October 13th** **In-session:** 1. **Q & A / Admin** (5-10 min) - Address questions 2. **`tidyr` & `dplyr` Tutorial** (45 min + 10 min Q&A) - data wrangling using `tidyr` and `dplyr` - Cleaning data in a script - Merging datasets - See: tidy+dplyr [introduction]( and [tutorial]( - _Suggested resources are still listed above in Session 4_ 3. **Lecture: Jennifer Sunday** (20-25 min) - Example work flows for data synthesis in meta-analysis and data aggregation, with a focus on reproducibility - See: [session 5 lecture slides]( * * * #### **Session 6: Thursday, October 15th** **In-session:** 1. **Q & A / Admin** (5-10 min) - Address questions 2. **Lecture: Laura Pollock** (35-45 min) - Random and mixed effects models - Assignment 2 3. **Activities** (30 min) - Discussion from lecture - Discussion: When to do a meta-analysis vs. a primary data aggregation **Post-session follow-up:** - Check out this tool that [simply explains hierarchical modelling]( (with graphics!). - Looking for more information on nested and crossed random effects? You can find a short summary [here]( - This is a really nice overview of [mixed modelling in R](, including random intercept and slope models, and random effects structure. **Assignment** (post-session): - **Option 1**: Develop a short research proposal for a future data synthesis project that builds upon your graduate research project, and includes your proposed question, statistical approach, and a summary of the available data. - **Option 2**: Bring together data and perform an analysis using meta-analysis or primary data aggregation techniques. Prepare a short report that motivates your question(s) and summarizes your results. - **Due date**: Please upload your assignment to your personal OSF Submissions component **by the end of Tuesday, November 3rd**. - **Rubric**: We've created a [rubric]( for each option. - **What is a research proposal?**: read more [here]( * * * #### **Session 7: Tuesday, October 27th** **Pre-session readings:** - Sunday, J., Bennett, J.M., Calosi, P., Clusella-Trullas, S., Gravel, S., Hargreaves, A.L., *et al.* (2019). Thermal tolerance patterns across latitude and elevation. *Philosophical Transactions of the Royal Society B*. [PDF]( **In-session:** 1. **Q & A / Admin** (5-10 min) - Address questions 2. **Activities** (50-60 min) - Tutorial: fitting a mixed effects model to aggregated data - See: [introductory lecture]( and [modelling aggregated data tutorial]( * * * #### **Session 8: Thursday, October 29th** **Pre-session readings:** **Suggested readings:** - Book: [Data Analysis Using Regression and Multilevel/Hierarchical Models]( **In-session:** 1. **Q & A / Admin** (5-10 min) - Address questions 2. **Lecture: Laura Pollock** (25-30 min) - Misc. things to consider as you move forward with meta-analysis/data agg. - Data visualization - Jenn - Predictive ability (sensitivity analysis) - Laura Bring it back to Nakagawa et al.’s (2017) ten appraisal questions for meta-analysis - Future of data synthesis - Laura/Jenn 3. **Activities** (45 min) - Open Q & A with students * * *
OSF does not support the use of Internet Explorer. For optimal performance, please switch to another browser.
This website relies on cookies to help provide a better user experience. By clicking Accept or continuing to use the site, you agree. For more information, see our Privacy Policy and information on cookie use.

Start managing your projects on the OSF today.

Free and easy to use, the Open Science Framework supports the entire research lifecycle: planning, execution, reporting, archiving, and discovery.