Continuing my posts on methods often used in cleaning data, this one discusses Appending data from two sources to make a single dataset.
The Append operation attaches new cases to the bottom of existing variables. To append one variable to the tail of another, choose the top variable as Y (click on its icon, for example), and then the variable to append to it as X (Shift-click on its icon, for example). Then choose
Manip > Append and make Group Variable
You’ll get a new variable (named “Data”) and a group variable (named “Group”), both placed in a new Relation.
Wait. What is this “Group” variable?
Well, that’s part of DD doing things differently. The group variable repeats the names of the variables that have been appended, labelling each case with the variable it came from. This can be very handy. For example, this is the form of data you’ll need for the ANOVA and General Linear Model functions, which want the data and the group names as arguments. You can also use the group variable to undo the Append with the
Manip>Split into variables by Group
Function. Choose the appended data as Y and the group names as X.
The appended data are placed in a new Relation because they have a different number of cases. They thus no longer link to the original relation.
One more thing: The Append function is not limited to appending one variable to another. Read the next posting to see how it deals with several variables.