Database Reference
In-Depth Information
In this chapter, we'll cover the simplest of quantitative comparisons: single variables as
either measurements (how much) or counts (how many). In later chapters, we'll focus on ele-
ments such as time and location, and then we'll explore how to best apply these techniques
to fields such as statistics, finance, and data journalism.
Communicating “How Much”
To start with, let's consider comparisons involving quantities that are measured rather than
counted. This includes fields such as revenue, weight, distance, and time, among countless
others. Most likely, your data source doesn't include a record for each dollar, pound, mile, or
hour. Rather, a single record could be a transaction like a sale or a shipment, each having a
measurement associated with it:
▪ A sales order resulting in revenue of $95
▪ An overnight shipment weighing 5.2 lbs.
▪ A flight covering 2,408 miles in 5 hours and 28 minutes
Additionally, your data source could be a summary table, such as total monthly revenue for a
number of months, or shipping weight for all packages. In this format, each row in the data
set is an aggregation of a number of individual records. It's important to understand the level
of aggregation of your underlying data set.
You already know from Chapter 1 that these measurements form quantitative data types, as
opposed to ordinal or nominal. And you know from Chapter 2 that Tableau sees them as con-
tinuous Measures (green pills), like the Area sq-mi field for each of New York's boroughs.
An Example of How Much
Staying with the New York theme in our example data, let's consider garbage. How much
garbage does the City of New York Department of Sanitation (DSNY) collect from each bor-
ough? Well, each borough is further divided into community districts. A summary table of
the total tonnage collected from each of the 59 community districts during September 2011 is
available on the NYC Open Data site .
Figure 3-1 shows a portion of the data table.
Search WWH ::




Custom Search