Lecture: Tuesday and Thursday 1:30-2:45 ENLAB 248 Zoom
Credits: 3
clark.rushing@usu.edu
Office: NR 146 Zoom
Office hours: Monday and Wednesday 1:00-2:30 (or by appointment)
Philosophical advantages
Philosophical advantages
1) Probabilistic treatment of all unknown quantities
Philosophical advantages
1) Probabilistic treatment of all unknown quantities
2) Coherent framework for incorporating prior knowledge into analysis
Philosophical advantages
1) Probabilistic treatment of all unknown quantities
2) Coherent framework for incorporating prior knowledge into analysis
3) Proper accounting of uncertainty
Philosophical advantages
1) Probabilistic treatment of all unknown quantities
2) Coherent framework for incorporating prior knowledge into analysis
3) Proper accounting of uncertainty
4) Ease of estimating latent variables (and uncertainty)
Practical advantages
Practical advantages
1) Ability to develop custom/complex models to suite your needs
Practical advantages
1) Ability to develop custom/complex models to suite your needs
2) Many statistical concepts (e.g., random effects) make more sense (no blackbox)
Practical advantages
1) Ability to develop custom/complex models to suite your needs
2) Many statistical concepts (e.g., random effects) make more sense (no blackbox)
3) Expanded "toolkit" for quantitative analysis
Practical advantages
1) Ability to develop custom/complex models to suite your needs
2) Many statistical concepts (e.g., random effects) make more sense (no blackbox)
3) Expanded "toolkit" for quantitative analysis
4) Ability to keep up with the literature
Practical advantages
1) Ability to develop custom/complex models to suite your needs
2) Many statistical concepts (e.g., random effects) make more sense (no blackbox)
3) Expanded "toolkit" for quantitative analysis
4) Ability to keep up with the literature
5) Ability to review manuscripts/proposals that use Bayesian methods
Disadvantages
Disadvantages
1) Computationally intensive
Disadvantages
1) Computationally intensive
2) Few "canned" software
The use of Bayesian methods is growing rapidly in ecology, largely due to:
The use of Bayesian methods is growing rapidly in ecology, largely due to:
The use of Bayesian methods is growing rapidly in ecology, largely due to:
The use of Bayesian methods is growing rapidly in ecology, largely due to:
These developments have led to many people adopting Bayesian methods even if they don't fully understand what they're doing. They also led to some ecologists becoming full-on Bayesians and then a predictable backlash from those that view these methods as needlessly complex and overly trendy (i.e., statistical machismo)
The methods you'll learn about in this class, just like all statistical methods, are tools to help you answer questions
Your job as a researcher is to choose the tools that best suite your question and your data. The goal of this course is to expand your toolbox
What makes research "reproducible"?
Many definitions (Goodman et al. 2016) but for the purposes of this class, we will define it as:
The ability of independent researchers to reproduce scientific results using the original data and methods (adapted from Markwick et al. 2018)
What makes research "reproducible"?
Many definitions (Goodman et al. 2016) but for the purposes of this class, we will define it as:
The ability of independent researchers to reproduce scientific results using the original data and methods (adapted from Markwick et al. 2018)
What makes research "reproducible"?
Many definitions (Goodman et al. 2016) but for the purposes of this class, we will define it as:
The ability of independent researchers to reproduce scientific results using the original data and methods (adapted from Markwick et al. 2018)
Goodman, S. N., Fanelli, D., and Ioannidis, J. P. A. (2016) What Does Research Reproducibility Mean? Science Translational Medicine, 8, 341ps12–341ps12.
Marwick, B., Boettiger, C. & Mullen, L. (2018) Packaging Data Analytical Work Reproducibly Using R (and Friends), The American Statistician, 72:1, 80-88, DOI: 10.1080/00031305.2017.1375986
1 Stricly speaking, reproducing results could also include the ability to recreate figures, tables, and even text as it appears in the original report or paper. Fully reproducible research is very hard but that does not mean we shouldn't strive for moving our work closer to to that end of the spectrum.
Making your work reproducible involves extra work. Why bother?
Making your work reproducible involves extra work. Why bother?
1) To help advance science - if your work can't be reproduced, it's not science
Making your work reproducible involves extra work. Why bother?
1) To help advance science - if your work can't be reproduced, it's not science
2) To meet requirements of journals/granting orgs
Making your work reproducible involves extra work. Why bother?
1) To help advance science - if your work can't be reproduced, it's not science
2) To meet requirements of journals/granting orgs
3) To make it easier to share your work with collaborators
Making your work reproducible involves extra work. Why bother?
1) To help advance science - if your work can't be reproduced, it's not science
2) To meet requirements of journals/granting orgs
3) To make it easier to share your work with collaborators
4) To make it easier to revise your analysis later
You always have at least one collaborator on every project - you future self. And your past self doesn't respond to email
Cooper, N. & Hsing, P. (2017) A guide to reproducible code British Ecological Society
WILD6900: Computational Tools for Reproducible Science (Dr. Simona Picardi)
Lecture: Tuesday and Thursday 1:30-2:45 ENLAB 248 Zoom
Credits: 3
Keyboard shortcuts
| ↑, ←, Pg Up, k | Go to previous slide |
| ↓, →, Pg Dn, Space, j | Go to next slide |
| Home | Go to first slide |
| End | Go to last slide |
| Number + Return | Go to specific slide |
| b / m / f | Toggle blackout / mirrored / fullscreen mode |
| c | Clone slideshow |
| p | Toggle presenter mode |
| t | Restart the presentation timer |
| ?, h | Toggle this help |
| Esc | Back to slideshow |