diff --git a/docs/group-project-page.md b/docs/group-project-page.md index 9b4ccbf..391cb7d 100644 --- a/docs/group-project-page.md +++ b/docs/group-project-page.md @@ -10,6 +10,8 @@ teaching team. **Important Note:** Although this is a group project, some related assignments will be submitted individually. You can (and are encouraged to) discuss it with your group members. However, *every* student will submit their own assignment and will receive an individual grade. +**Submission:** All assignments must be done in a Jupyter notebook, and then submitted both as an `.html` file (`File` → `Download As` → `HTML`) and an `.ipynb` file that is reproducible (i.e. works and runs without any additional files). + ### Deliverable 1: Team Contract (1% weight) A **group contract** is a document to help you formalize the expectations you have for your group members and what they can expect of you. It will help you think about what you need from each other to work effectively as a team! You will create and agree on this contract as a team. **Each member should "sign" (you can just type out your name) at the bottom of the submission**. At a minimum, your group contract must address the following: @@ -35,12 +37,10 @@ How will we address non-performance regarding these goals, expectations, policie Many data analyses start with a question or questions we are intrested in. Then we find or collect data to answer the question(s). Decide with your team members a topic of interest you want to work on and identify a dataset to work with. Include a brief description of the dataset (2 or 3 sentences) and a link or reference on how to access the dataset. Note that in the next individual assignment you will be asked to fully describe the dataset. -### Individual assignments +### Deliverable 2-4: Individual assignments (total 12% weight, 4% each assignment) Although this is a group project, the following 3 assignments related to a dataset chosen by the group will be submitted individually. You can (and are encouraged to) discuss them with your group members. However, you don't need to come with a common solution and *every* student has submit their own assignment and will receive an individual grade. -All assignments must be done in a Jupyter notebook, and then submitted both as an `.html` file (`File` → `Download As` → `HTML`) and an `.ipynb` file that is reproducible (i.e. works and runs without any additional files). - #### Assignment 1: Data and Question(s) *This is an individual assignment*. Every student needs to write and submit their own assignment. You must submit **two files** in Canvas: @@ -74,84 +74,52 @@ In this assignment, you will: **Note**: this visualization does not have to illustrate the results of a methodology. Instead, you are exploring which variables are relevant, potential problems that you anticipate to encounter, groups in the observations, etc. -It is fine to have idea as other group members. However, you don't need to agree on a unique common visualization for the group project. In fact, usually the exploratory data analysis will have many different visualizations! Regardless of how many plots are proposed within each group, *each team member* needs to propose one visualization and justify their choice. - -#### #### Assignment 3: Methods and Plan - -The previous sections will carry over to your final report (you’ll be -allowed to improve them based on feedback you get). Begin this *Methods* -section with a brief description of “the good things” about this report -– specifically, in what ways is this report trustworthy? - -Finish this section by reflecting on how your final report might play out: - -- What methods do you plan on using? -- What do you expect to achieve? -- What impact could your results have? - -#### References - -At least two citations of literature relevant to the project. The citation -format is your choice – just be consistent. Make sure to cite the source -of your data as well. - -### Deliverable 3: Peer Review - -For this peer review, you will provide feedback to another proposal. While this is an individual assignment, I encourage you to work in group so that you are all explosed to multiple proposals. However, each of you need to submit comments to the student assigned to you. +It is fine to share ideas with other group members. However, you don't need to agree on a unique common visualization for the group project. In fact, usually the exploratory data analysis will have many different visualizations! Regardless of how many plots are proposed within each group, *each team member* needs to propose one visualization and justify their choice. -#### Instructions +#### Assignment 3: Methods and Plan -To complete the peer-review assigned to you, write comments in Canvas' text box or through an attached file. Recall that you'll be evaluated in the comments and feedback given. See the rubric added to the Peer Review assignment in Canvas. **Do not enter a grade!** +Propose one method to address your question of interest using the selected dataset and explain why it was chosen. In your explanation respond to the following questions: +- Why is this method appropriate? +- Which assumptions are required, if any, to apply the method selected? +- What are potential limitations or weaknesses of the method selected? -Remember to be _respectful and provide constructive comments_. +As mentioned before, it is fine to share ideas with other group members. However, you don't need to agree on a unique common method for the group project. In fact, usually the analysis comprise different methods with different strengths and limitations! Regardless of how many methods are proposed within each group, *each team member* needs to propose one method and justify its choice. -Following a recommendation to reviewers of BMC Genomics Journal: "Before you submit your report, please take a moment to read it through and put yourself in the place of the authors. How would you feel if you received this report? Would you find the tone courteous and professional?" Are your comments useful to strength other groups' analysis? +### Deliverable 5: Interview with your TA (4% weight) -**General guideline** -Your review should contain the following elements: +The whole group will meet for 5 minutes with the TA to suggest a plan on how to combine all the material in a final report. Students will be graded individually. Your grade will be based on your participation in the interview and the quality of your arguments and explanations. -- As you read through the proposal, point out anything that you think is confusing, or is not -communicated effectively. When possible, provide suggestions for improvement. If everything looks -good to you, say why it looks good. -- What part of the proposal is the most effective, and why? -- What part of the proposal is the least effective, and why? Provide a suggestion for improvement. -- Provide feedback on English, spelling, and grammar (if applicable). +### Deliverable 6: Individual assignment (4% weight) -The rubric can be found on Canvas. In short, *mechanics* (10%) evaluates -the composition of your submission, *reasoning* (70%) evaluates your -feedback, and *writing* (20%) evaluates your English, spelling, and -grammar. There is no page limit. +This will be the last individual assignment. Write a computation code to implement the method proposed in the previous assignment (or suggested in the interview) and use a visualization or table to report the results. In 3 or 4 sentences give a brief interpretation of the results. If needed, comment on any unexpected result or potential problems with the analysis, and possible ways to address issues encountered. If results are as expected, explain how they address the question of interest. -### Deliverable 4: Final Report +### Deliverable 7: Group Final Report (5% weight) -Each group will create a final electronic report (max 2000 written -words, not including citations) using Jupyter to communicate the +Each group will create a final electronic report (max 2000 written words, not including citations) using Jupyter to communicate the question asked, the analysis performed and the conclusion reached. -Only one member of your team needs to submit. You must submit **two files**: +Only one member of your team needs to submit. The group will work on combining the material from all individual assignments. The way in which the material is combined will depend on each project. You don't need to include everything done. Just make sure that all team members agree on the final content of the report. Discuss options with your TA and Instructor if needed. + +You must submit **two files**: - the source Jupyter notebook (`.ipynb` file) - the rendered final document (`.html` file) -- Each report should include the following sections: - - Title - - Introduction - - Methods and Results - - Discussion - - References + +Each report should include the following sections: + - Title + - Introduction + - Methods and Results + - Discussion + - References #### Introduction -Just be sure to improve this section by incorporating feedback, and changing -things based on your own improved understanding of the project. - Begin by providing some relevant background information on the topic so that someone unfamiliar with it will be prepared to understand the rest -of your proposal. +of your proposal. -Identify and describe the dataset that will be used to answer the -question. Remember, this dataset is allowed to contain more variables -than you need, in fact, exploring how the different variables in the dataset affect your model, -is a crucial part of the project. +Propose one or two questions you want to examine and describe the dataset that will be used to answer the +question(s). Also, be sure to frame your question/objectives in terms of what is already known in the literature. Be sure to include at least two @@ -160,10 +128,9 @@ include these in the References section). #### Methods and Results -#### Preliminary Results - -In this section, you will: +In this section, you will include: +a) "Exploratory Data Analysis (EDA)" - Demonstrate that the dataset can be read from the web into R. - Clean and wrangle your data into a tidy format. - Plot the relevant raw data, tailoring your plot in a way that addresses your question. @@ -173,27 +140,16 @@ In this section, you will: Be sure to not print output that takes up a lot of screen space. - - -Here is where you’ll include your work from the “Preliminary Results” in -your proposal, along with the additional results you planned to conduct, -as indicated in the “Methods: Plan” section of your proposal. Be sure to -incorporate feedback from the teaching team and your peers (as -relevant), or make any improvements based on your own improved -understanding of the project (now that more time has passed since the -proposal). - -Specifically, in addition to what is requested in the “Preliminary -Results” section of the proposal, we are looking for the following -components: +b) “Methods: Plan” - Describe in written English the methods you used to perform your -analysis from beginning to end that narrates the code the does the +analysis from beginning to end that narrates the code that does the analysis. -- the "Feature Selection" process, how and why you choose the covariates of your final model. +- If included, describe the "Feature Selection" process, how and why you choose the covariates of your final model. - Make sure to interpret/explain the results you obtain. It’s not enough to just say "I fitted a linear model with these covariates, and my R-square is 0.87". - if inference is the aim of your project, detailed interpretation of your fitted model is required, as well as a discussion of relevant quantities (e.g., are the coefficients significant? how is the model fitting the data)? - a careful model assessment must be conducted. + - if prediction is the aim of the project, describe the test data used or how it was created. - Ensure your tables and/or figures are labeled with a figure/table number. #### Discussion @@ -205,12 +161,13 @@ In this section, you’ll interpret the results you obtained in the previous sec - Discuss how your model could be improved; - Discuss future questions/research this study could lead to. - #### References -The same instructions for your proposal also applies here. You only need to make changes if necessary (e.g., if feedback indicates so). +At least two citations of literature relevant to the project. The citation +format is your choice – just be consistent. Make sure to cite the source +of your data as well. -### Deliverable 5: Team Evaluation +### Deliverable 8: Team Evaluation (2% weight) Evaluate each member of your group (including yourself) in terms of how they/you participated, prepared, helped the group excel, and was a team player. @@ -231,4 +188,4 @@ Click the Teammate Evaluation Template link in the Canvas home page to access th ### Attribution -These instructions are a modified version of the DSCI 100 and STAT 201 project at UBC. +Part of these instructions come from descriptions of the DSCI 100 and STAT 201 projects at UBC.