The final project involves conducting a rigorous data analysis using statistical approaches that we have learned during the semester (or other tools as approved by the instructor). You can choose whichever statistical methods are suitable to your project scope and questions, but they must be methods we cover in this class (unless otherwise approved) and appropriate to your data. Please turn in your own work and assignment, but you may use as many resources as you wish (class material, online material, or the insight/experience of other students) to help generate the results and figures.
The write-up will loosely take the format of a scientific paper to be submitted to a professional journal. However, because of the nature of this course, the most important pieces of the write-up are the methods and results sections. Nonetheless, I expect at least a few paragraphs introducing the topic and why it’s important, and a few paragraphs discussing the implications of the results. Be concise- include only enough text to convey the important information for each section! See below for details.
I encourage you to talk with your advisor to see if there is a way you can make the final project directly part of your thesis project or at least relevant to your thesis/dissertation project. You will end up spending quite a bit of time on your final projects, so think about what types of analyses you want/need to get to know better as part of your graduate research. If you don’t have a dataset to work with you can choose to work with a public dataset- there are lots of great datasets out there… see links page
Please provide a 1-page project description. Make sure you include:
Here is a more detailed description of expectations for the manuscript.
Title: Choose a concise yet informative title. Puns are acceptable!
Author: Although you will likely have more than one author if/when you publish this work, all submissions should have only one author: you!
Abstract: Please include a short abstract (200 words max.) that summarizes your work (context, research questions, results, and conclusions).
Introduction: Provide enough description so that the reader understands why the research is important and (if appropriate) what research question(s) and/or hypotheses are being addressed. (approx. 2-3 paragraphs). Always keep the big picture in mind!
Methods:
Results: Present all relevant results completely yet concisely. Wherever possible, results should be presented via figures and tables. There is a limit of 4 figures and 2 tables, so choose carefully which figures and tables to present. Figures should be publication quality, with well-labeled, readable axes. Make sure figure captions have enough information for a reader to interpret the figure without reading the main manuscript! Remember that each figure can contain multiple panels so you should have no trouble fitting all your key results into 4 figures max.
Discussion: Write at least two paragraphs that put the results in a larger context (returning to the key research questions) and discuss areas of uncertainty. Potential topics are possible violations of assumptions, and future work that your analysis suggests would be profitable.
Acknowledgments: Please acknowledge key people and agencies that helped with data collection, funding, analysis, providing comments, etc.
Literature cited Full references (exact format up to you, but be consistent!) for all citations in your main text. Make sure you cite key R packages that you use- and make sure to cite R. The ‘citation()’ function in R makes it easy to know how to cite R and R packages.
Supplement Provide all code used to run the analyses presented in the paper as an R script. Only present the code for running the analyses you present in the paper- you don’t need to include all side tangents and alternative visualizations, analyses you did not present, etc. You are welcome to use a github link instead.
NOTE: please embed figures and tables within your main text- you do not need to put them in a separate section at the end of the manuscript.
NOTE: you may also choose to include additional (supplemental) tables and figures at the end of your document. But such figures and tables will not affect your final project grade.
NOTE: please include line numbers- this is important for the peer-review process
Your first drafts will be subjected to peer review.
Unlike standard scientific peer-review practice, the peer-reviews in this class will be not be anonymous– please sign your reviews.
Your manuscript will be reviewed by two other students, and the reviews themselves will be graded (see below; one peer review document should be turned). I have placed an example peer-review in WebCampus.
The reviews are designed to provide feedback on all aspects of the paper- but for this class the most important aspect of the peer reviews feedback on the statistical approach being used- and suggestions for improvement!
Key questions for peer-reviewers to consider are (in no particular order):
The peer review assignment is an individual assignment, with two submissions due per person (one review for each paper reviewed).
Each submission should be approximately two pages in length and should include:
NOTE: keep your comments respectful and constructive (of course!). And you can point out things you really like about the paper as well!
NOTE: you don’t need to correct grammar and spelling here- just focus on the big stuff!
I will collate the reviews and submit the peer-reviews to each author along with my comments (comments from the ‘editor’!). This is analogous to how you will receive feedback after submitting to a journal.
I look forward to reading your final papers!