Content area
Full Text
photo, Stephen Few
Grid lines are commonplace in business graphs today but they are almost always "chartjunk" - visual content that adds no value, serves no purpose and distracts from the real data.1 However, there are a few exceptions. In this column, I'll describe the few times that grid lines actually help graphs to communicate.
Most software that supports the creation of graphs automatically includes grid lines as the default format, but it ought to be the other way around. You should need to intentionally add grid lines in those uncommon circumstances when they actually serve a purpose. Unlike data tables, graphs are not meant to provide precise quantitative values. Graphs reveal patterns, trends, relationships and exceptions via the shape of the data that would be difficult to discern from a table of values. Grid lines are rarely needed in graphs to help readers assign accurate numeric values to the data; the approximate values that can be perceived without the aid of grid lines are almost always adequate.
Quantitative and Categorical Scales
Graphs are constructed using axes - lines that serve as boundaries for the space in which the data is displayed, along which scales appear to assign meaning to the data. Two major types of scales appear in graphs: quantitative and categorical. Quantitative scales assign numeric values to equidistant locations along an axis. They enable you to interpret the quantitative values in the graph - encoded as bars, lines or data points - by matching the position of these visual objects to a marked position along the scale line. Categorical scales assign labels to positions along an axis to identify what is being measured, such as sales regions, business departments and so on. Most graphs contain two axes called the X- axis (horizontal) and the Y-axis (vertical). Usually one provides a quantitative scale and the other a...