# Linking Data Sets

NOTE: This page contains information about a future version of Rave. Your version can't do this yet.

# Introduction

Two data sets may be "linked" when each contains a variable that represents the same information, encoded with identical values. For example, a data set of Sales Records and a data set of Product Inventory may both contain a variable of Product ID Numbers. In this case, the two data sets may be linked by their respective Product ID Number variables such that filtering on this variable in one data has the same effect as if the variable were simultaneously filtered in BOTH data sets.

Links may be either one-way or two-way. A one-way link from data set A to data set B causes data set B to update in response to a change to data set A, but data set A will not update in response to changes to data set B. A two way link causes either data set to update in response to a change to the other. One way links are always defined between exactly two data sets, but two-way links can exist between any number of data sets, such that changing one data set updates all other data sets that share in the two-way link.

In order to link data sets, you must first have both data sets loaded into Rave. The variables in each data set that will be used to define the link do not have to have the same name (though for clarity you'll probably want them to), but they do need to contain values encoded in exactly the same way, and the variables all must be the same data type in Rave. For example, if one variable in a link is discrete, the corresponding variable in the other data set to be linked must also be discrete.

Links can be created and modified from the Manage Data Sets GUI.