The field preview provides context-aware metadata about your fields to help you understand the relationships in your data model and get some quick stats about the data and fields.
Key fields link your tables together across common values. Clicking on a key field will show you the total distinct values, some example values, and all of the tables in which this key field appears. In the example below, the Country field appears in 2 tables.
There are two important metrics that Knarr also surfaces here: density and subset ratio. Density is how populated the field is, i.e. if you have 156 total records and 100% density, that means that all 156 records contain values (non-null) in that field.
Subset ratio describes what percentage of total field values are represented in this table, i.e. “TOTAL COUNT” divided by “DISTINCT VALUES”. Typically, for dimension tables, this should be at or close to 100%.
Non key fields will provide similar information as key fields, but will also include breakdowns of common values and whether or not the field has duplicate values. In the example below, we can see the most common Age is 23, representing 8.1% of the total records in this table. The table also has duplicate values, which is expected as a dimensional field.