Sometimes, you may need to quickly rename a variable, change a column name, add a description to a variable or make other changes to dataset metadata. In this case, it may be easier and quicker to do so directly in the platform without altering to the dataset file itself. There is a possibility to edit dataset metadata directly in the platform through the Dataset Viewer. This function is available for users who have edit access to a dataset.
This article describes what types of metadata are available for editing in the platform and how to work with them.
Which types of metadata are available for editing?
Dataset metadata can be classified into the following categories:
- Dataset-wide metadata, including:
- Dataset name;
- Dataset description;
- Data source (provider);
- Original URL;
- Terms of use and other legal fields (data citation, license, etc.);
- Tags (topics);
- Publication and next release dates;
- Update priority;
- Replacement dataset.
- Dimension-level metadata, including:
- Dimension (column) names;
- Dimension element (category) names;
- Descriptive fields (e.g. variable description, ID, etc.).
Dataset-wide metadata appear in the Details tab on the left panel of the Dataset Viewer on in the footer if the the window size is narrow.
Dimension-level metadata is available by clicking "i"circle near the indicator name in the right panel of the Dataset Viewer.
How can I edit dataset metadata?
All these metadata can be accessed and edited throught the Dataset Viewer > More actions > Edit details.
To rename a dimension or its elements scroll down to the section "Dimensions" and find needed dimension.
Note: we call a column that contains categories a Dimension. In the platform, there are two special types of dimensions that have specific mandatory metadata fields, they are Region, and Company. Region-type dimesnion has "Region ID" field which is used to identify regions across the platform. Company dimension has "Ticker" field that uniquely identifies listed companies across datasets.
To make changes to dimension metadata, click its name. Also, you can re-order dimensions by dragging and dropping them in the list.
Here, you can rename a dimension itself or its elements, add or remove a descriptive field and fill this field with some information.
How to add a new metadata field to a dimension?
You may need to add a new metadata field, for example, when you have a dimension "KPI" and want to provide a definition of each metric so that users of the dataset can get a sense of these metrics.
To add a new metadata field, navigate to the Dataset Viewer > More actions > Edit details > Dimensions. Go to a required dimension and click Add button.
Define a field name and type and save. Next, you would want to fill this field with information for each dimension member. To do that, click a dimension member, fill the metadata fields with information, and save.