Skip to Main Content
It looks like you're using Internet Explorer 11 or older. This website works best with modern browsers such as the latest versions of Chrome, Firefox, Safari, and Edge. If you continue with this browser, you may see unexpected results.

Research Data

A guide for locating, managing, and sharing research data.

Introduction to Data Visualization

Why is data visualization so important?

Biomedical data are rapidly increasing in both volume and complexity. Given the importance of these data in clinical practice it is imperative that they are communicated clearly and effectively. However, faulty processing of data, especially visual data, accounts for upwards of half of all misdiagnoses. Good data visualization can help ensure biomedical data, which are often costly and difficult to acquire, are interpreted correctly by practitioners.

Anscombe's Quartet is a set of two-dimensional data which are identical in their calculated mean, variance, and correlation coefficient. However, these data plot very differently! This is one example of the importance of visualizing data.

 

Best Practices

1. Use color sparingly, and be mindful of color choices.

2. Order your variables.

Your labels or data groups should appear in a logical order, be it alphabetically, by size, or some other meaningful listing.

3. Use the correct visualization for your data type(s).

  • For example, when using continuous data consider replacing bar graphs with a more informative visualization, like dot plots

4. Make image based figures clear and informative.

Include clear scale bars and magnification insets where appropriate. Use colors that will translate well to grayscale, when possible. Ensure your images are high enough resolution for the situation and have not loss quality during submission for publication.

Additional Resources

Bibliography