Why visualize your data?
By Kevin Chan Ting
This article may have been translated automatically. Therefore, it can contain gramatical mistakes and imprecisions. We’re looking for solutions and will update as soon as possible.
Here is a throwback to my years in university., During these years, among other things, I studied economy and statistics. Obviously, I had quite a few research papers to write (such great memories! read as sarcastically as you can) and data to analyze.
To share my analysis, I would describe data with a few paragraphs and some very basic graphs. Those Excel graphs were, to that point, my only contact with data visualization. Actually, the terms « data visualization didn’t mean much to me, even if we are confronted to it every day.
Why am I writing about all this ?
Because today, I no longer am a student but a trainer at Yoono. Because today, I realize how much my capacity to handle data has evolved.
Be mindful of the method used to share your results, its just as important as the analysis it self. If you cannot communicate your results effectively, why waste your time and energy ?
To illustrate this article, we will use a case study. It will show what data visualization can bring.
I receive data from the Angers Open Data website to analyze. (https://data.angers.fr/pages/home/).
I am asked to present an analysis of Angers university’s student body according to departments (displayed as “composante” in the Excel sheet), gender, nationality and this, over the course of several years.
Here is the type of questions, my analysis should be able to bring answers to:
is the evolution of the number of students?
According to gender? Nationality? Department?
there departments with a severe gap between the number of male students vs.
Departments with more foreign students than others? All that filtered by year…
Data analysis enables you to answer questions…
Visual analysis enables you to answer faster!
WITHOUT data visualization
Let’s take the second question as an example. As a student, I would have displayed my data as an Excel sheet to then compare all the values of all departments. After that, I would have written a text to explain my interpretation of the Excel sheet.
Does this Excel sheet help you to easily answer the question about parity? The number of foreign students compared to French students? You’ve got 30 seconds!
It is still quite easy to measure the number of female students compared to the number of male students, for each department, studied separately. Problems start to arise when you want to compare among departments or compare ratios… All the sudden it’s a lot more complicated.
WITH data visualization
This is where visualization becomes extremely useful. What does it bring?
When used correctly, it allows to save time, answer rapidly to questions and sometimes, even discover trends or imperceptible behavior when reading your dataset. The core idea is to make data more accessible and more attractive.
With data visualization, discover trends or imperceptible behaviors in one glance!
Here under, is an example of the graph visualization of the data contained in the previously mentioned Excel sheet:
Isn’t it easier to display the distribution of all students for each department and compare them between one another?
On the one hand, the simple Excel sheet means reading the numbers, one after the other. Therefore, you have a segmented analysis for each item.
On the other hand, the grath allows a much more synthetic vision of the information.
The idea is not to oppose the two methods (Excel sheet and data viz) but rather to use them in a complementary manner. The graph allows a global view, the identification of behavior or trends. Then the use of the Excel sheet is useful when you need to analyze in details, a specific element.