PSY 652: Research Methods in Psychology I

Introduction to Data Visualization

Kimberly L. Henry: kim.henry@colostate.edu

What is a visualization?



“A visualization is a representation designed to enable exploration, discovery, or communication.” – Dr. Alberto Cairo

The cover of Alberto Cairo's The Art of Insight

Why create visualizations?

The data science pipeline.

Why create visualizations (the data science pipeline)?

The data science pipeline, with additions for how visualization comes into play.

Why start with data visualization?

Good Reason #1

Decorative image of fuzzy monsters painting plots.

Artwork by @allison_horst

Good Reason #2

Decorative image of a brain.

Good Reason #3

Decorative image of going from R struggles to R success.

Artwork by @allison_horst

The INCREDIBLE POWER of data visualization

Great potential for harm (part 1)

WAPO headline: Why this National Review global temperature graph is so misleading.

Great potential for harm (part 2)

National review tweet of a graph downplaying change in global temperatures by vastly expanding the y-axis.

Great potential for harm (part 3)

A graph showing that small changes in average temperature have big impacts on civilization.

Great potential for harm (part 4)

Two side by side graphs showing average global temperature with a very wide y-axis and a more reasonable y-axis.

Two charts from the WAPO article mentioned previously.

But also, a great potential for positive impact

Variations of Earth’s surface temperature:

The famous hockey stick graph showing change in average global temperature, with a large. uptick after the industrial revolution.

Michael Mann, Ph.D., a Climate Scientist, created this graph in 1999. It is believed by many to be one of the most influential charts ever produced.

Alberto Cairo’s 5 key principles of an effective visualization

The “hockey-stick” chart embodies Alberto Cairo’s 5 key principles of an effective visualization

The cover of Alberto Cairo's book: The Truthful Art

  1. It is truthful, as it’s based on thorough and honest research.
  2. It is functional, as it constitutes an accurate depiction of the data, and it’s built in a way that lets people do meaningful operations based on it (seeing change in time).
  3. It is beautiful, in the sense of being attractive, intriguing, and even aesthetically pleasing for its intended audience—scientists, in the first place, but the general public, too.
  4. It is insightful, as it reveals evidence that we would have a hard time seeing otherwise.
  5. It is enlightening because if we grasp and accept the evidence it depicts, it will change our minds for the better.

Truthful

Visualizations should present data accurately, avoid misleading the audience, provide the best obtainable truth.

Functional

Visualizations should convey the intended message, allow the audience to interpret it easily, should require minimal effort to understand.

Beautiful

Visualizations should be aesthetically pleasing, making the information accessible and enjoyable to explore.

Insightful

Visualizations should reveal new insight.

Enlightening

Visualizations should enlighten the audience.

Do these charts meet the criteria?

To access the survey, press and hold control, click the link below, and choose “Open Link in New Tab”.

Go through the survey with your neighbor. Discuss each chart – then enter your own rating.

Click here to take the survey.

A step by step guide to a great chart by Cole Nussbaumer Knaflic

The steps

The cover of Knaflic's book: Storytelling with Data

  • Understand the context
  • Choose an appropriate display
  • Eliminate clutter
  • Draw attention where you want the audience to focus
  • Think like a designer
  • Tell a story

An example

Decades of rising income inequality and slowing economic growth have significantly eroded a core element of the American dream: the belief that each generation will fare better than the one before.

A fivethirtyeight headline: Inequality is Killing the American Dream

Our goal: To visually illustrate the dramatic decline in economic mobility in the United States over time. That is, to highlight how, over the past several decades, the likelihood of children earning more than their parents has sharply decreased, signaling a significant shift in the economic landscape.

Step 1: Understand the context

Begin by identifying your audience and their needs. What is the key message you want to convey? Consider what data you have available and how it can support the story you aim to tell.

The audience and key message

  • Audience: Students, educators, policymakers, and possibly a general public interested in economic and social issues. They may have varying levels of familiarity with the underlying data and research methods but share a common interest in understanding the factors contributing to economic mobility and inequality.

  • Audience Needs: The audience needs a clear, compelling visualization that highlights the decline in economic mobility and its implications. They are looking for a visual that not only presents data but also helps them grasp the broader narrative of how the American dream is changing and what that means for future generations.

  • Key Message: The key message to convey is that economic mobility in the U.S. has significantly declined over the past few decades, primarily due to rising income inequality (a smaller proportion of people own the majority of income/wealth). This decline has profound implications for the American dream, as fewer people are able to surpass the economic success of their parents.

Data available

The available data includes trends in absolute economic mobility across different birth years (cohort), highlighting the proportion of 30-year-olds (age30_absmob) and 40-year-olds (age40_absmob) who earned more than their parents at the same age. The data were collected by Dr. Raj Chetty and colleagues of Opportunity Insights.

Step 2: Choose a display

Select the type of chart or graph that best represents your data. Whether it’s a bar chart, line graph, scatter plot, or something else, the choice should make the data easy to interpret and relevant to the audience.

A bar chart 🙁

A bar chart of the probability of earning more than your parents by birth cohort

A line graph 🥰

A line chart of the probability of earning more than your parents by birth cohort

Step 3: Eliminate clutter

Simplify your visual by removing any elements that do not add value. This includes unnecessary gridlines, labels, or colors that might distract the viewer from the main message.

Remove the legend, annotate instead

Line graph with legend removed.

Step 4: Draw attention where you want the audience to focus

Use design techniques like contrast, color, or annotations to highlight the most important parts of your chart. This ensures that the audience’s attention is drawn to the key insights.

Bold lines and text, use a minimal background

Line graph with bolding and minimal background.

Step 5: Think like a designer

Pay attention to the overall layout, alignment, and spacing in your chart. Ensure that it is visually appealing and easy to read. Consistency in design choices can make your chart more professional and understandable.

Left align the title and try a different font

Line graph with better aligned title and interesting font.

Step 6: Tell a story

Finally, craft your chart to convey a narrative. The data should guide the audience through a beginning, middle, and end, ultimately leading to a clear and compelling conclusion.

Addition of narrative elements

Line graph with annotations to tell the story of the data.