Trellising

Filtering is very powerful as it allows you to filter out values from the dataset so that you can focus on what's important at any particular point. However, it's not very useful for comparing different subsets of the data. In the preceding example, we were comparing histograms by marking (selecting) different parts of the data and filtering out various rows of data. We can look at much more at a single glance by using trellising. Trellising splits visualizations into panels so that you can see subsets of the data all at the same time.

I suggest trellising by pclass—this is so that we can compare the survival rates of passengers in the various classes on board the Titanic:

1. Reset the filters as described in Step 6 in the Using filters section.
2. From the data panel, click the pclass column and drag it onto the age histogram. You'll see a pop-up panel appear:
   
3. Drop pclass onto Use 'pclass' on the vertical trellis axis.:
   

1. Notice what happens to the visualization (presuming we have selected all the data in the main bar chart):

1. We have some fresh insights!
   • Unsurprisingly, but rather unfortunately, your chance of survival was much better as a first-class passenger than if you were from any other class. Notice the 80-year-old first-class passenger all on his own.
   • Third-class passengers were much more numerous than second-class ones, but their survival rate was much, much worse than the other classes'.
   • All second-class children survived.

1. Experiment with marking different subsets of the data in the main bar chart and observe how the trellised histogram behaves - see if you can gain additional insights!