About

You will be assigned two review the draft group project for two peers. Projects should be submitted to canvas as a link to a git repo and a list of files worked on by that specific group member.

Please remember to be respectful and constructive in your comments. The purpose of peer review is twofold:

Responsibilities

As a peer-reviewer, you are responsible for reviewing the following areas:

  1. core and additional analyses for the group member you were assigned;
  2. sections common to all group members: introduction, data, methods, results and discussion;
  3. git utilization.
  4. professionalism and quality of the presentation.

In reviewing the code used for the analyses (1) please follow the existing peer review guidelines. It is also helpful if you briefly compare to another group member to verify that the analyses are parallel with differences clearly explained and justified.

A rule-of-thumb for the expected effort of these peer review is that you spend ~30 minutes on each of your two reviews.

Additional guidelines

Below are some additional guidelines on structuring your peer review for responsibilities 2-4.

Common Sections

Here, it would be helpful to address:

  1. Does the project contain the required sections?
  2. Is the substantive question clearly presented and easy to identify?
  3. Are the data and methods adequately described?
  4. Do the data and methods descriptions match what is actually done in the analyses?
  5. Is the writing style clear and concise? Please highlight any areas you find confusing or difficult to follow.

Git Utilization

Here, it would be helpful to address:

  1. Is there a readme that describes the project and the relation between files?
  2. Are all source files present and organized in a helpful way?
  3. Is there evidence of multiple commits per member and some form of code reviews?

As a group, you may wish to explain in the readme your evidence of collaboration.

Quality of presentation

Here, it would be helpful to address:

  1. Is the project as a whole easy to follow and navigate?
  2. Does the project have a professional appearance?
    • Do snake_case or other coding conventions appear in titles, figure axes, or other places they donโ€™t belong?
    • Do figures and tables have brief explanatory legends?
    • Are model results summarized in a nice way?

Nitpicks

Here are a few things I will nitpick for points, please help your peers avoid them.

  1. Source scripts: missing headers or header information, inadequate comments, and inconsistent styling.

  2. Inadequate or unclear evidence of collaboration using git.

  3. Presentation quality:
    • excessive decimals or other lack of attention to formatting,
    • confusing use of NHANES variable names rather than the quantities they represent,
    • lack of confidence intervals for (key) point estimates.