Applied Quantitative Methods
Structure data, build models, and plot results in the R programming language
Class projects may cover any topic that involves quantitative methods. Take a look at the page of links to datasets which could be analyzed as part of a project.
Students are expected to contribute:
Graduate students are also required to contribute:
The project code will be submitted via your personal project specific github repository. If your repository is private you will need to invite your instructor to be a collaborator so that they can examine the code and test it out. The home directory of all projects should contain at least the following directories:
All R code in the scripts directory must assume that the working directory is the project home directory and all file paths must be relative to the project home directory. Points will be deducted for absolute file paths as these decrease the portability and readability of code.
Do not commit very large > 100 MB data files to the git repo. Instructions for how to download these files or other justifications for why the data are not included with the code are sufficient.
If the data are not available to reproduce the results then at minimum a representative example portion of the data must be included to provide a means of generating example results.
The project directory should also contain a README.md file that describes (at a minimum):
Although not required your instructor and your future-self will find it very useful if you include a master script that controls project flow. See for example https://github.com/weecology/mete-spatial/blob/master/ddr_run_all.R Another very effective approach is to use an Rmarkdown document that walks a reader through you analysis with $ code and results interspersed with plain English descriptions of motivations and methodology. See for example http://richfitz.github.io/wood/wood.html
A 10 minute presentation accompanied by slides on your:
The oral presentation should summarize the broader context within which your work falls by citing the peer-reviewed literature. It should be clear what your over-arching question is and what specific questions you have attempted to address Your data and statistical methods need to be adequately described. We do not need to know which R packages or what R code you used but we do need to know the names of the methods you used and how you examined your hypotheses. Some projects will not use data and thus that portion can be skipped in those contexts.
At a minimum this should include:
However those that wish to tackle an entire scientific paper are encouraged to do so and your instructor will give you comments on your entire document. The sections of the written description should be formatted and prepared in the style of a relevant scientific peer-reviewed journal in your field that you would like to submit the finished product to. Scientific literature should be cited in the methods and interpretation sections of the document.