Assignment #4 (HW #4)

Assigned Mon 01/06/2025 | Due Sat 03/08/2025

Learning Outcomes

Your final assignment is meant to combine all of the course learning outcomes(!):

  • identify which types of visualizations are most appropriate for your data and your audience
  • prepare (e.g. clean, explore, wrangle) data so that it’s appropriately formatted for building data visualizations
  • build effective, responsible, accessible, and aesthetically-pleasing visualizations using the R programming language, and specifically {ggplot2} + ggplot2 extension packages
  • write code from scratch and read and adapt code written by others
  • apply a DEI (Diversity, Equity & Inclusion) lens to the process of designing data visualizations
  • assess, critique, and provide constructive feedback on data visualizations

Description

Choose to complete one of the below options:

OPTION 1: Infographic

Build a cohesive infographic-style visualization that includes at least three different, but complementary visualizations that work together to tell a complete story. You should have one overarching question with (at least) three sub-questions which are addressed using each of the component pieces of the infographic.

Infographics are useful science communication pieces, which can serve a variety of audiences, but are particularly well-suited for engaging general audiences. An infographic typically exists to address one overarching question or idea, and it’s subcomponents (e.g. visualizations, numbers, imagery) each help to tell part of the story (or in other words, help to answer sub-questions). While text is important in all visualizations, it can be particularly critical in weaving together all of your infographic elements to successfully convey your story. The order and orientation in which your elements are positioned will also be essential for creating a visual hierarchy that successfully guides your readers. Packages like {patchwork} will likely be critical in stitching together the different elements of your infographic.

A few examples from Winter 2024 submissions:

And a few more fun examples from other {ggplot2} creators:

OPTION 2: Three viz, same question, different audiences

Build three separate visualizations, each to serve a different target audience / purpose. Each visualization should use the same data to answer the same question, but how you choose to present those data should change based on the following:

  1. A visualization for technical writing / subject matter experts: The purpose of this visualization is to clearly communicate key insights, trends, or patterns, and is one that could be published in a technical report or academic paper. These types of visuals may include more details and domain-specific language. They should, however, still prioritize clarity by finding ways to highlight the most important takeaway message(s). Two great examples are Michael Mann’s infamous “Hockey Stick” graph, and Figure 1 (Unusual climate anomalies in 2023) in Ripple et al. 2023.

  2. A visualization for policy- or decision-makers: The purpose of this visualization is to balance clarity and insight with actionable and evidence-backed data to influence or inform policy decisions. A challenge is effectively communicating key insights, trends, forecasts, comparisons, etc. while remaining accessible to non-experts. These visuals often leverage clear, catchy titles, minimize potentially distracting design elements, present only the most essential data, and use text annotations to draw attention to key areas of a the graphic / help tell a story. A great example is Figure SPM.1(c) (Adverse impacts from human-caused climate change will continue to intensify) from the Intergovernmental Panel on Climate Change (IPCC)’s Climate Change 2023 Synthesis Report (see page 7).

  3. A visualization for public outreach / a general audience: The purpose of this visualization is to raise awareness, prompt discussion, or otherwise engage an audience who may not have domain-specific knowledge or expertise. These types of visuals should limit (or fully omit) the use of any jargon. They may display findings in an atypical but creative way, as well as blend both art and science. A great (and famous) example is Ed Hawkins’ Warming Strips

OPTION 3: Propose an alternative project idea

I want this assignment to serve you and your goals – if you’re not excited about Options 1 or 2, you may choose to propose an alternative idea. Please submit a brief (should not exceed 1 page) proposal at the time of your HW #1 submission. Your proposal must include / address:

  • a clear description of your proposed project
  • how you plan to incorporate at least three distinct data visualizations
  • the intended audience(s) of your data visualizations
  • any software or tools you’ll need to use and / or learn (beyond the necessary data wrangling / plotting packages, e.g. {tidyverse})
  • how this alternative project option serves your growth in the field of data visualization and communication

Please be aware that proposing an alternative project idea does not guarantee that it will be accepted.

Regardless of which option you choose, you are expected to consider all design elements discussed throughout the quarter, implement as appropriate, and justify your decisions. These include, but are not limited to:

  1. graphic form (you are not limited to just those fundamental chart types discussed in weeks 2 & 4 – explore other advanced chart types and don’t be afraid to get really creative with it; check out some of these awesome data viz creators to find inspiration)
  2. text (e.g. titles, captions, annotations, axis labels, axis ticks, alt text)
  3. themes (i.e. all non-data plot elements; these should be intentionally modified and visually-pleasing)
  4. colors
  5. typography
  6. general design (e.g. group order, spacing, text orientation, data-ink ratio, creating a visual hierarchy, avoiding information overload)
  7. contextualizing your data
  8. centering your primary message
  9. considering accessibility (e.g. colorblind-friendly palettes / contrast, alt text)
  10. applying a DEI lens to your design, as appropriate (e.g. considering the people / communities / places represented in your data, consider how you frame your questions / issue)

In addition to producing your visualizations, you are expected to follow clean and organized coding practices:

  • follow the tidyverse style guide (important styles are outlined on the course website)
  • organize and annotate code (see these recommendations)
  • apply the appropriate code chunk options (e.g. in most cases, code and outputs should be rendered but warnings and messages should not) – see the Quarto documentation on HTML Code Blocks and Execution Options for more
    • (optional) I recommend also applying the code-fold: true option, which collapses code chunks and makes it easier for readers to focus on your writing and embedded visualizations (while still allowing them to explore your code)

You will also share your data design process and decisions in a short, 1-2 page (~500-1,000 words) science communication blog post-style write up. Your writing should:

  1. include an engaging introduction paragraph(s), which states your motivation, question(s), data / data source(s), and goal(s) of your visualization. A few great examples:
  2. address your approach and decisions for each of the ten design elements listed above (though you are welcome and encouraged to comment on any others that are not explicitly listed). If your visualizations do not include or consider a listed element(s), please comment on why (it’s possible that not all ten elements will be applicable to your visualizations).
Storytelling helps to engage your readers

Be sure to review the EDS 296 materials on writing data science blog posts for tips on crafting a story and engaging your readers!

You may submit your assignment in one of two ways:

  1. As a blog post on your personal website – check out the Adding a blog to your existing Quarto website materials for a refresher, if needed
Remember, this is an opportunity to grow your professional portfolio!

Regardless of which of the two publishing options you choose, consider storytelling strategies for effective and engaging data science writing – review the Blogging for data science slides from the EDS 296-1F course materials for guidance. Remember, this is an opportunity to grow your portfolio and showcase both your technical and communication skills (through data viz and writing).

  1. A published Quarto document that is deployed using GitHub Pages
  1. Your Quarto doc must be named index.qmd and live in your repository’s root directory. Be sure to rename / move it, if necessary.
  2. Render index.qmd and ensure that index.html appears in your root directory as well (this is the file that will be deployed by GitHub Pages).
  3. Head to your remote repository on GitHub. Navigate to the Settings page (top navbar), then select Pages from the left-hand menu. Once there, you should see something that looks like this:

  1. Under Build and deployment > Branch, update the drop down that says None to main. Leave the second drop down on /(root) (this tells GitHub to look for and deploy our index.html file from the root directory of the main branch). Click Save. Your URL will appear at the top of the page once it’s deployed (you will need to refresh the page – this could take a few minutes):

Finally, you are expected to present your final data visualization product to the class.

You will have 3 minutes to present your final data visualization(s). You must address the following in your presentation:

IMPORTANT:

These are in-person presentations, held during finals week. Make-up presentations will not be permitted. Please plan accordingly!

Rubric (specifications)

You must complete the following to receive a “Satisfactory” mark for Assignment #4: