In this article we show how elements get their data values and the different attributes you can edit to change an element's data source.
Initial data values
By default each element looks for a data column with a column name equal to the key of the element. For instance, element Q1 will get is data from data column Q1.
Field attribute
You can specify a different data column using the field attribute. For instance element Q1a_recoded's field is set as Q1a. When you clone an element the field is automatically set as the key of the original element. If the original element itself had a different field, then the clone would have the same field.
You can specify more than one data column for field. For instance, Q7 has its field attribute set to an array: Q7_1, Q7_2. This can be an effective, lightweight way to combine multiple data columns into a single element.
Element attribute
You can specify another element as the data source using the element attribute. This takes the final post-calculation value from element as the starting value for another element. This can be an effective way to create hierarchical codes, using the coded values from one element as the input to a higher level code frame.
The values in Q7_recoded2 are the already recoded values from Q7_recoded. I then recoded the existing codeframe into broader sentiment categories.
Transform attribute
You can use the Calculate/Transform dialog to create a new value for parent elements.
A parent element is a group of elements instead of a single element.
For Q7 we can choose to transform the group to "Condense" which combines values from both child elements (Q7_1, Q7_2) into the parent's distribution. This is similar to setting field to an array of values except heritable properties apply to children but not fields.
