Suggested journal papers and additional research papers as mentioned in references have provided a new insight into data visualization and gave a new direction on how I can incorporate the changes in my day to day reports, Below I explained why I feel Data visualization is important to me and how the suggested papers gave me a new direction.

Why is Data Visualization important?

Data Visualization is an important part of the data analytics world. My experience comes from working with large finance data, I first build my dashboard which included general bar graphs, pie charts, and drill downs. I used tools like Dundas and tableau to get to my final dashboard(Example below).

The role of data visualization is to get useful insights from the data so that senior management can make valuable decisions from it. The process of building a dashboard included different steps

  1. Understanding the Data

The first step before building a dashboard is to understand the data you are working with. Spend time with the data to understand the correlation of the data points. Understand the final objective.Perform Univariate analysis to understand if there are any outliers in the data. Build time series plots to understand the data flow. To complete this exercise tools like Excel, R or SAS can be used.Below is a example financial dashboard

  1. Understanding your Audience/Stakeholders

The second step is to understand the final stakeholders/audience/ end users and how technical the dashboard needs to be made. For Example if the dashboard is for Senior management then its preferable to show month over month comparison and Delta etc, as mentioned in the paper “Skau, Kosara - 2016 - Arcs, Angles, or Areas Individual Data Encodings in Pie and Donut Charts”,ArC Length, Donut charts and pie charts are good visual cues. The below figure shows a complicated dashboard with too much information which is not effective. This can be solved by the next step creating mock dashboards.

  1. Creating multiple mock dashboards

Creating multiple mock dashboards helps the stakeholder and the creator to get a better understanding of the data. The below example shows a refined dashboard which is straight forward and also easy to understand.

  1. Refining the dashboards by sharing it with shareholders

Based on the feedback from the stakeholders the dashboard needs to refined and needed controls have to be implemented to see that the right data is flowing through the dashboard. The creator also needs to keep in mind the frequency of usage of the dashboard. Most of the dashboard contains Line graphs, Bar charts with dynamic filters. A line graph is invented by William Playfair in 1786 to help people understand time series data.

  1. Maintaining the dashboard

Once a dashboard is built the end user or the creator should take responsibility for maintaining the dashboards and reviewing the controls and upgrading the dashboard as needed.

References

W. Javed, B. McDonnel, and N. Elmqvist. Graphical Perception of Multiple Time Series. In IEEE Transactions on Visualization and Computer Graphics, Volume: 16, Issue: 6, Nov.-Dec. 2010.

Skau, Kosara - 2016 - Arcs, Angles, or Areas Individual Data Encodings in Pie and Donut Charts

J. Heer, N. Kong, and M. Agrawala. Sizing the horizon: the effects of chart size and layering on the graphical perception of time series visualization. In Proceedings of the ACM CHI 2009 Conference on Human Factors in Computing Systems, pages 1303-1312, 2009.

W. S. Cleveland and R. McGill. Graphical perception: Theory, experimentation, and application to the development of graphical methods. Journal of the American Statistical Association, 79(387):531-554, September 1984.

J. Heer and M. Bostock. Crowdsourcing graphical perception: Using mechanical turk to assess visualization design. In ACM Human Factors in Computing Systems (CHI), pages 203-212.

J. Talbot, V. Setlur, and A. Anand. Four Experiments on the Perception of Bar Charts. In IEEE Transactions on Visualization and Computer Graphics, Issue: 12, Dec. 31 2014.