The type specification in a channel encoding describes the type of data attribute being mapped to a visual channel. This is a required parameter for the automatic vis specs inferrence of DXR to work. Currently, DXR supports the following data types:

• quantitative - numerical values that describe some form of quantitative data, e.g., weight of a car in tons, temperature at a given time in degrees Celsius.
• nominal - categorical data in which the data attribute has a discrete set of possible values; in this case there is no sense of order among the different categories. Examples of nominal data are: names of countries, names of collaborators, types of cars, origin country of cars, keywords in a book.
• ordinal - is a form of categorical data in which there is a natural order to the discrete set of possible values though the distances between values cannot be quantified. Examples of ordinal data are: size - small, medium, large; difficulty level - easy, intermediate, difficult; Likert scale - like, like somewhat, neutral, dislike somewhat, dislike.

To get a better sense of these data attribute types, please check out the example below and other examples provided in the Examples page.