Interpreting Data Visualizations: The Basics: Home - Research Guides

Skip to Main Content

1. ZSR Library
2. Research Guides
3. Interpreting Data Visualizations: The basics
4. Home

Search this Guide Search Interpreting Data Visualizations: The basics: Home Charts, graphs, and more. Tips on how to read them. Examples of common data visualization mistakes. Purposes of different graph & chart types.

Purpose

• This guide has been created because all data visualizations, whether a chart, graph, infographic, etc, should be read with a grain of salt. Data is misinterpreted more than you may think. Even with the best of intentions, visualization creators may omit, over-simplify, or over-complicate important variables.
• As a reader of data visualizations, your goal is to understand, interpret & reflect on the information represented & then infer new information based on the assessment. This can be difficult if you're not familiar with data or statistics.
• The tips below are broad enough to apply to any kind of data visualization scenario.

Tips for reading charts, graphs & more

1. Identify what information the chart is meant to convey.

Remember, data visualizations are created with a purpose in mind: to be used as evidence to support an idea or claim. What is that idea or claim?

2. Identify information contained on each axis.

3. Identify range covered by each axis.

4. Look for patterns or trends.

Do see any clusters, steady increases or decreases, consistent coloring?

5. Look for averages and/or exceptions.

6. Look for bold or highlighted data.

7. Read the specific data. Data is typically read as [number of Y's per X].

8. Look for citations for the data to see where the data originated and to ensure it is credible.

• Ask your self the following questions about the where the data originated:
• Who created the data?
• Is the creator an expert in this area?
• When was the data created?
• What geography does it cover?
• Is it objective data (just the facts)? Or, is there some kind of bias?
• Does the data make sense? Do you believe it?

More Visualization Examples & Tips

• Information is Beautiful Examples of beautiful data visualizations!
• Spurious Correlations Examples of correlations gone wrong
• WTF Visualizations Examples of terrible data visualizations
• Calling Bullshit Videos from UW's Information School's course, Calling Bullshit. This course focuses on identifying inaccurate data in research & in the media.
• A Quick Guide to Spotting Graphics That Lie National Geographic's tips to spotting visualizations that lie.
• The Chart that Tim Cook Doesn't Want You to See Example of Apple's misleading cumulative graph from 2013.

Common (but not all) Visualization Mistakes

1. Truncated (shortened) Y Axis aka "broken scale"

In most cases the Y Axis ranges from 0 to the maximum value that encompasses the range of the data at hand. However, sometimes the range can be changed to better highlight differences in the data. BUT, when taken to the extreme, this technique can make differences seem much larger than they actually are.

2. Misleading Cumulative Graphs (showing data that is increasing in quantity)

Beware of cumulative graphs. For example: Instead of showing a graph of quarterly revenue, we could display a running total of revenue earned to date. In this case important variables may be omitted, over simplified, or over complicated.

3. Ignoring Conventions

Violating standard practices of visualizations. For example: pie charts that represent parts of the whole and timelines that progress from left to right. This is a problem because we are wired to misinterpret this data due to our reliance on these conventions.

4. Failed Calculations

For example: pie charts should add up to 100%. This can happen when survey takers can select more than one response.

5. Visualization type is all wrong

When the creator of the visualization chooses a chart type based on esthetic taste rather than the character of their data.

6. Displaying too much data

This causes the visualization to be overcomplicated and causes the visualization to be near impossible for the audience to figure out.

7. Trying too hard to be original

Recently infographics and graphical representations of data have become popular due to the fact that most people, despite culture, read charts in the same way. Due to this fact, sticking to conventions is crucial to audience understanding. Arrange data intuitively (alphabetically, sequentially, or by value) and in a logical way.

8. Making the reader work too hard

Some visualizations alone are not enough and require the creator to add qualifying numbers, text or trend lines.

9. Purposeful & Selective bias

• Purposeful bias: deliberate attempt to influence data, most likely to take the form of data omissions or adjustments
• Selective bias: slightly more discreet and passes by those who do not, or are not able, to read between the lines, such as: the nature of the sample of people surveyed. Example: Surveying college students about legal drinking age

10. Using percentage change in combination with a small sample size

Alongside the choice of sample (see #9), an additional factor to be aware of is the size of the sample. When an experiment or study is led on a totally not significant sample size, not only will the results be unusable, but the way of presenting the results as percentages will be misleading.

Example: Asking a question to a sample size of 20 people where 19 answer "yes" is a 95% "yes" answer rate versus asking the same question to 1,000 people and 950 people answer "yes" giving a 95% "yes" result rate again. Validity of the percentage is clearly not the same.

12. Correlation implying causation

Just because two sets of numbers follow a similar path doesn’t mean there’s a correlation.

13. Ignoring population size makes accurate comparisons impossible

If the visualization is talking about where people live, it is important to note how many people live there.

14. Decoration can be distracting & misleading

Beware of altered perceptions due to 3D modeling

Free & Easy Data Visualization Tools

• Google Data Studio Did you really think there wouldn’t be a Google product on this list? Google has their own data visualization program, Data Studio, that’s free, and of the quality you’d expect from Google. Also, since it’s Google, all their properties (AdWords, YouTube) can connect with Data Studio. If you’ve already got a Gmail account (or any Google account), setup with Google Data Studio is as easy as a few clicks. If you’ve already got a Google account, it’s also probably the easiest to get started with of all the products on this list. In keeping with Google’s usual for-dummies design, all you have to do to start playing with a visualization is copy one of the templates they provide for you (unless you want to upload your own data).
• Tableau Public Tableau Public is the business intelligence giant’s free offering, and it doesn’t disappoint. You get up to 10 GB of storage, the program has a drag-and-drop interface, and you get up to 15 million rows per workbook. The downside is that you can only save to your public profile, so maybe avoid that visualization with potential leads, yearly revenue, or everyone’s blood type (I’m sure some company, somewhere, does that, and I don’t judge).
• Open Heat Map Creating a map is simple. You upload a CSV, Excel, or Google Sheets file to Warden’s website, click a button and get a map. It’ll live on his site, but you can link to it. This is all to say that Openheatmap is about as simple as sending an email with an attachment.
• Palladio Palladio is specifically designed for history teachers, but is also useful for businessmen. Palladio has a simple drag-and-drop interface, and can convert CSV, TAB, or TSV files. The range of visualizations you can make is short (maps, galleries, graphs and lists), but if you’re looking to make something like a map of customer locations, or a gallery of customers with info about them (see pic for example), Palladio could be useful.

Hoppe, Geoff. 2017/06/07. 22 Free & Open Data Visualization Tools. Retrieved from https://blog.capterra.com/free-and-open-source-data-visualization-tools/.

Analysis & Assessment of Data VIsualizations

1. Consider other factors that may have shaped the data & therefore the visualization.

What factors not measured in the data set could have affected how the data is represented?

2. Consider the creator (bias & authority check)

• Who did the primary research?
• What did the scientist or statistician try to figure out?
• How big was the sample size? Who was apart of the sample? How inclusive was it?

3. Reflect & Interpret

What is the takeaway of the visualization based on patterns and other factors?

4. Infer further

What other information can you reason based on this interpretation?

Librarian

Summer Krstevska Contact: [email protected] 251C Farrell Hall336-758-4931 Subjects: Accounting, Business - General, Economics, Entrepreneurship, Finance, Marketing & Advertising

More from ZSR

• Ask ZSR
• Citation Guides
• Printing in ZSR
• Zotero Citation Manager
• Zotero Workshops
• ZSR Delivers & Interlibrary Loan
• How Do I . . . ?
• ZSR Research Guides & Tutorials
• WFU Writing Center
• ZSR Library Calendar
• ZSR Events!

• Last Updated: Sep 23, 2024 10:35 AM
• URL: https://guides.zsr.wfu.edu/interpretdataviz
• Print Page

Login to LibApps Subjects: Social Sciences - General Tags: charts, data literacy, data visualization, graphs, statistics, visual literacy Report a problem.