Traditional Table vs. Modern Flow Map : A quantitative comparison between two information visualizations

Abstract: In today’s world we have access to endless amounts of data of varying kind. We would like to be able to display data in a way that actually feels intuitive and not intimidating. The field of information visualization and data visualization has been around for about 40 years attempting to overcome this very problem, and we are still learning more and more about how to visualize different types of data today. One major problem is that there are so many types of data that it is very difficult to make generalizations about how to display certain types of data. In our study we are comparing two specific visualizations with the purpose of wanting to find out if we can draw any types of general rules or guidelines from doing so. The two visualizations we are comparing are a flow map and a tabular visualization depicting interstate migration between US states. Both visualizations show the exact same data but the flow map is a more graphically pleasing representation of the data whilst the tabular visualization looking more like a classic spreadsheet. We have done this by letting 48 people fill in a webform with multiple choice questions. Each participant was randomly assigned either the flow map visualization or the tabular visualization as their aid in filling out the form. Our participants were mainly KTH students and also students that found our study on facebook and deciding to participate. In the past a few qualitative studies have been conducted researching effective methods for visualizing information. However, there is a clear lack in quantitative studies made as far as we know. Therefore we decided to conduct this study attempting to measure the two visualizations performance and compare them against each other. A qualitative part was also included in the web form where we asked our subjects about their experience filling out the form. Our results shows that the flow map outperforms the tabular visualization with regards to both accuracy, efficiency, and perceived difficulty. In a case similar to our study we can conclude that a flow map seems to be the right choice over a traditional tabular visualization. However, one has to bear in mind the context of the study is rather specific. A context where the visualization has other purposes than finding specific answers to simple questions might yield other results. Our main contribution is in methodology, in that we define and implement a low footprint study of interactive data visualisation.