You are viewing an old version of this page. View the current version.

Compare with Current View Page History

Version 1 Next »

Joined data sets allow you to analyze data from more data sets together and therefore use data from more data sources in a single report. This function might help you to compare internal company data to external - business intelligence then easily becomes competitive intelligence (CI).

The principle is similar to joining SQL database tables.

One data set could be joined with more data sets (eg. join by department ID to get department full name, address, country, total sales and then join by product ID to get products name, price, weight and mass). Each join is called join point.


Tutorial video

Join condition

Each join has following parameters:

  • Name: Joined data set name.
  • Join with: Name of the source data set to join with.
  • Compare type: Date/time in both records are taken in account or not. Note: Codebook records usually do not contain time information.
  • Join condition: Standard join types available - match always depends on attributes.
    • Left outer join: Record in the target data set is not mandatory.
    • Inner join: Record in the target data set is mandatory.
    • Cross join: No attribute match required.

The resulting joined data set will contain all the attributes and indicators from all source data sets.

  1. Data in the joined data set will be the intersection of all joins.
  2. Data in joined data sets are updated automatically when data are changed or imported to the source data sets. You cannot import data directly to the joined data sets.
  1. Please note, that the permissions setting based on data filter is not available joined data sets!
  2. You cannot join a data set with already joined data set.
  3. Data join is also available on data source level (database). Please consider this option when you plan to analyze millions records of data. Joining on database level may provide better performance.

Joined data set supports these functions the same way like in ordinary data set:

Next Steps

  • No labels