Interpret a Data Visualization

Posted on May 15, 2019 by Gustavo

Today is my first school day at the nanodegree Business Analytics program at Udacity, and the very first thing we had to do is to write a report interpreting data with Tableau. The data depicted some interesting facts about the city of Madrid. Tableau is amazing for data visualization and this is a good example. Next comes an important part of data science: the correct interpretation of the data. This was my take on the task.

BsAn pyramid

One of the features that most impressed me while interacting with the dashboard is how weird looking the population pyramid looks. Later, by focusing on the district distribution, I could note that, with some small differences among each district, the tree-like shape was common in all of them. Digging a little deeper on each district, I would notice that the common feature implies a lower population in the younger segment, especially among the ages 15 to 19 segment. We don’t know what causes this feature (although we could speculate with a couple of ideas), but it is a significant one.

BsAn votes n prices

It is interesting to notice that the political tendency of the typical Madrid citizen is well correlated with his/her amount of wealth, inferred here by the second-hand housing price. It can be noticed that citizen living in more expensive houses tend to vote for right-wing parties while citizen with less expensive houses prefer to vote for left-wing parties. It is also interesting to note that one district – the one named Centro- behaves fairly different.

BsAn cars n prices

As noticed in the precious insight, the Madrid district named Centro behaves differently than the rest. Although it can be found amount the wealthiest district (or at least, among those that show higher second-hand housing prices), it´s inhabitants prefer to vote for left-wing parties, opposite to the other wealthy districts. Moreover, when looking at the chart that shows the relationship between house prices and cars per 100 inhabitants, we can notice that it owns much less cars that the other wealthy districts. We can’t infer the cause directly from the data and it is probably an specific circumstance characteristic only to this district (maybe it’s is an old European district with narrow streets, making it very difficult to find proper parking, or maybe cars are not required as much because everything is at walking distance), but it’s definitely an interesting trait.