Recipe 2 – alternating between rows and records mode
Now let's have a look at how OpenRefine gives you access to multi-valued cells. When we follow the instructions of the previous recipe to split a column's values, we see that OpenRefine does two things. On one hand, it takes the first of the different values of a cell and puts it in the original row. On the other hand, it takes each of the remaining values and puts all of them in a cell of their own on an otherwise empty row. For instance, in the following screenshot, you can see that the record with ID 9 has been stretched out over three rows, each of which contains a category name. Only the first category is on a row that contains the other values of the field; the others are empty except for the category value (some columns have been hidden for clarity).
Note
A row is a single line of data in your dataset.
A record consists of all rows that belong to a single object. The first row of a record starts with non-null cells that identify the...