Draft specification for labels #45
Conversation
|
|
||
| ## Scope and specifications for labels and metadata | ||
|
|
||
| Adding support for labels in recodeflow will enhance data clarity and usability by providing descriptive metadata for variables, ensuring consistent interpretation across analyses. This feature will require implementing a standardized system for label management, including functions for adding, modifying, and retrieving labels within the recodeflow framework. There are two primary uses of variable and value labels: |
There was a problem hiding this comment.
adding, modifying, and retrieving to what? The variable and variable details sheet? The data?
|
|
||
| Other uses include: | ||
|
|
||
| 1. Facilitating data sharing and collaboration by providing clear documentation of variable meanings and valid values. |
There was a problem hiding this comment.
valid values, this isn't solved by labels right? Its solved by other metadata such as mins, maxes, units etc.
|
|
||
| 1. Facilitating data sharing and collaboration by providing clear documentation of variable meanings and valid values. | ||
| 1. Enabling automated report generation with human-readable variable descriptions. | ||
| 1. Supporting data validation by documenting expected values and ranges. |
There was a problem hiding this comment.
See comment above in point 1
| ```{mermaid} | ||
| graph LR | ||
| A[Raw Data] --> B[Labeled Data] | ||
| B --> C[Factor] |
There was a problem hiding this comment.
I don't understand the fork from B to factor and character. Do you mean categorical variables can be treated as a factor or a character variable?
| F --> G[Export] | ||
| ``` | ||
|
|
||
| Another significant challenge is maintaining label consistency across different representations of the same data. For example, a variable might need to be represented as both a labeled numeric vector and a factor, depending on the analysis context. Each transformation between these representations requires careful handling of both the values and their associated labels to ensure that the semantic meaning of the data is preserved. Additional challenges include handling missing values and their labels consistently, supporting multiple languages or label variants for the same variable, maintaining label integrity during data reshaping operations, ensuring labels remain synchronized when subsetting or filtering data, and managing the memory overhead of storing extensive label metadata. |
There was a problem hiding this comment.
Do you mind providing a concrete example of your generic example here?
There was a problem hiding this comment.
The additional challenges line can be removed, its in points form below.
|
|
||
| ## Label specifications | ||
|
|
||
| - Follow the labelled library approach. All labelled functions should work with recodeflow. |
There was a problem hiding this comment.
I would add a sub-section here about which challenges mentioned above recodeflow will tackle and which ones are out of scope.
The PR has specifications for refactoring labels to address the approach of moving labels out of rec_with_table(). Further, we have little use of labels and metadata, but this is a priority feature.
There is a clearer and more consistent approach to variable labels. I propose we should extend the 'labelled' library, and there are a few ways to do so.