Project Peer Review
Initial feedback due Oct 30 at 11:59p; Final feedback due Nov 5 at 11:59p
Critically reviewing others’ work is a crucial part of the scientific process, and STA 210 is no exception. Each group has been given read access to another group’s repo to review and provide feedback on their project draft. This review is intended to help you create a high quality final project, as well as give you experience reading and constructively critiquing the work of others.
To ensure the group has enough time to start incorporating feedback ,you should work on the peer review during lab and submit comments by Friday, Oct 30 at 11:59p. If you do not finish during lab, you can submit additional comments until Thu, Nov 5. Note that you are expected to submit the majority of the comments on Oct 29
Getting started
Click here to see which project you’re reviewing.
Then, search your GitHub repositories for project and you should see the repo for the project you’re reviewing.
Open the repo and follow the usual steps to clone into RStudio. You have read access to this repository but will not be able to push any changes.
Reviewing the draft
Carefully read the project draft, considering the questions below as you read it.
Once everyone in the group has read the draft, submit your responses to the peer review questions using “Issues” in the GitHub repo.
Post a new issue for each section of the peer review. In the issue, copy the peer review questions, and then type your responses. Every team member should post at least one issue.
To post an issue, go to the project repo on GitHub and click “Issues”. For each comment or critique on the written report and data analysis, click “New Issue”. Provide an informative issue title and then a description outlining your concerns, comments, and any suggested changes. Click “Submit new issue”. Do this for every major comment or critique (do not submit a single issue with all of your feedback).
Peer review questions
For each question - Copy the question in the GitHub issue and answer “Yes”, “No” or “Somewhat”. - If you answer “No” or “Somewhat” provide an explanation addressing any problems, i.e. why you didn’t respond “Yes”.
Click here for an .md file of the questions to make them easier to copy and paste in the GitHub Issue.
Issue 1: Introduction + Data
- Is the research question and goal of the report clearly stated?
- Does the introduction provide appropriate background context and motivation for a general reader?
- Is the source of the data stated with an appropriate citation?
- Is it clear when and how the data was originally collected?
- Are the observations and relevant variables described?
- Include any additional comments or suggestions on the introduction and data description.
Issue 2: Exploratory data analysis
- Is the data cleaning and data wrangling clearly described? This includes dealing with missing data, creating new variables, etc.
- Do the visualizations follow the guidelines we’ve discussed in STA 210? This includes using plots that are appropriate for the data, having proper axis labels, titles, etc.
- Are any tables and figures clear, effective, and informative? Should any be eliminated, or are any new tables or figures needed?
- Include any additional comments or suggestions for the exploratory data analysis.
Issue 3: Methodology + Results
- Do the regression methods match the research question(s) and available data? If not, point out areas for further work.
- Are the methods described in enough detail that the work could be replicated by someone else? Is it clear what approach and model were used to evaluate hypotheses of interest? If not, point out areas for further work.
- Is the analysis appropriately performed? What type of diagnostic methods were used to check any modeling conditions, and are you satisfied the conditions of the model are valid? Should any additional analyses be performed?
- Does the report contain a correct and effective interpretation of the results provided? Is all information needed to substantiate the results and conclusions included? If not, point out areas for further work.
Issue 4: Writing + Reproducibility
- Is the paper professionally presented and generally free of distracting errors or other issues, including (but not limited to) insufficient organization or formatting; poor grammar, spelling, or punctuation; or too-small font? Is the overall paper easily readable for someone with your expert level of knowledge? Note any concerns here.
- Knit the .Rmd file. Are you able to reproduce all aspects of the report, including output, visualizations, etc? This includes (1) being able to knit the document and (2) obtaining the same PDF as the original PDF.
- What questions and/or general feedback do you have for the authors?
- After giving feedback to this group, what is one thing you want to change or continue working on for your report?
Submission
The peer review will be graded on the extent to which it comprehensively and constructively addresses the components of the partner team’s report: the research context and motivation, exploratory data analysis, reproducibility, and any inference, modeling, or conclusions.
You will be graded based on the submitted issues on GitHub.