This project is due on Friday 10/16 at 11:59pm.
For this project, you will create an original data visualization using a dataset of your choosing. I want you to think of this as the beginning of a piece of work you might eventually include in a personal portfolio to demonstrate your data science skills.
Your goal is to make thoughtful decisions about what information to visualize and how best to represent it. You may need to clean, filter, or otherwise manipulate your data to accomplish this. You should be prepared to explain the decisions you made and why they are appropriate for your data.
Later in the semester, we may return to these visualizations to refine and improve them as you develop additional skills.
Your visualization must satisfy the following requirements:
ggplot2. Include appropriate
titles, axis labels, and legends (where applicable), customize the
appearance of your axes and/or legends, and use a deliberate color
scheme rather than the default colors.You may clean, filter, transform, or otherwise manipulate your dataset as needed to create your visualization. The goal is not to make the most complicated graph possible, but to create a thoughtful visualization that communicates something meaningful about your data.
You may work in groups of up to three students. Choose your group members thoughtfully. Working with friends can be enjoyable, but you should also consider whether your schedules, work habits, and expectations for the project are compatible.
All group members are expected to contribute meaningfully to the project and participate in the decisions involved in creating the visualization. Group work should be collaborative rather than a division of the assignment into separate pieces completed independently.
Before committing to a group, discuss your availability and make sure you can identify times when you will be able to meet and work together. You should not assume that you will be able to complete the project entirely through text messages or by exchanging files. If you are unable or unwilling to meet outside of class, I recommend working alone.
Each group will submit both their R Markdown
(.Rmd) file and a PDF generated from that
file. Your R Markdown document should contain the names of all
members of your group, and all code necessary to prepare your data and
produce your final visualization, along with a brief written discussion
addressing the following:
Keep your written discussion concise. The purpose of this assignment is to create a thoughtful, well-designed data visualization, not to write an extensive report. Your explanations should demonstrate that you understand your data and can justify the decisions you made.
Your PDF should display your final visualization clearly, with appropriate titles, labels, legends, and formatting.
Your project will be evaluated based on three broad criteria:
These are broad evaluation criteria rather than a checklist. Meeting the minimum technical requirements does not necessarily mean that a visualization is effective.
Reproducibility: Your submitted R Markdown file must contain all code necessary to reproduce your visualization from a fresh RStudio session. Your code should not depend on objects or commands previously run in your R environment. If your code requires external data files, you must include those files with your submission. A visualization that cannot be reproduced from your submitted files will receive a 10% deduction from the project grade.
Individual Accountability: All students are expected to understand the work submitted for this project, regardless of how responsibilities were shared among group members or what outside resources were used. You may be asked to explain or defend any aspect of your project, including the code, graphical choices, data manipulation, and interpretation. If you are unable to demonstrate an adequate understanding of the submitted work, your individual project grade may be reduced, even if other members of your group receive full credit.