Introduction to the data panel

You'll find that a lot of work is done in Spotfire via the data panel. You can show the data panel by clicking on the big icon in the middle of the Spotfire window, or by pulling it out by clicking on the data icon on the left-hand side of Spotfire:

The data panel shows all the data tables and columns that are available in the analysis. Spotfire has already classified the columns into different groups of numerical and categorical columns:

In this particular dataset, some categorical columns have been loaded as numeric columns. It's not Spotfire's fault—it's just that some of the data columns are integers in the data, and represent categories. Think of the column called survived. This is a 1 or 0, indicating whether the passenger, died or survived. Similarly, passenger class (pclass), the class of passenger should be categorical since it is either 1, 2, or 3, and taking any kind of aggregation of this (average, max, and so on) probably doesn't make much sense. You can read more about the dataset and its data dictionary here:
http://biostat.mc.vanderbilt.edu/wiki/pub/Main/DataSets/titanic3info.txt

Or:

http://bit.ly/2uDg6oX

Next, we are going to use Spotfire's recommendations engine to build a visualization, but in order to get the best results from it, it can sometimes be a good idea to change the categorization of columns in order to give Spotfire some hints about how to display or analyse the data. So, let's do this first:

1. Right-click the pclass column and change its categorization to Categories:

1. Do the same with the survived column.
2. Now, we can get started with building visualizations!