Definition
Structured information that explains everything about data, and it has several types (Descriptive, Administrative, Structural, and Markup Metadata).
Reference: https://www.quora.com/What-is-metadata-in-a-database
How to record
Metadata should be stored alongside your main data sets. This documentation should be maintained along with your active data sets.
- Readme files
- Data dictionary
- Protocol or methods
- Notebook documentation
Data Dictionary
A data dictionary defines and describes the elements of a dataset so that it can be understood and used later. If you are a data steward, you should be updating a data dictionary to align with the updates being made in the dataset. The information being collected here should be:
- Data Steward
- Contact Information
- Description of dataset
- Data source
- Data type
- If data is spatial, provide a Coordinate System as well
- Time range of data
- Update Frequency
- Explanation of variables
- Format
- Database connection
- Access Level
- Primary Key
Template
A basic starting point to create a metadata dictionary. You can customize this template according to your specific requirements and fill in the details for each data element within the respective columns of the table.
| Element Name | Description | Data Type | Source System | Owner | Usage | Dependencies | Data Length/Precision | Index/Keys | Updated | Notes |
| Customer_ID | Unique identifier for a customer | Integer | CRM System | Customer Service | Customer profile | NA | NA | Primary Key | 2024-04-03 | NA |
*Future link to downloadable starter template