/AGGRE <data> BY <aggr_variable> TO <new_data_file> / K.Vehkalahti
combines observations in a Survo data file according to values
of <aggr_variable>. Only active variables will be processed and
saved to <new_data_file>. Also <aggr_variable> must be active.
The <data> is automatically sorted according to <aggr_variable>.
If <new_data_file> already exists, it is overwritten.
Sucro /AGGRE itself is a combination of existing Survo modules,
namely FILE operations SORT, AGGR, COPY, STATUS and UPDATE,
offering an easy way for the most typical applications. For more
demanding tasks, FILE AGGR includes several additional options.
The observations having the same value in <aggr_variable> are by
default combined (aggregated) by computing the means. However,
aggregation may be controlled by a specification AGGRE having
the value of SUM, MEAN, MEDIAN, MIN, MAX or NMISS.
The observations are then combined either by summing the values,
computing the means, medians, minimum or maximum values or by
summing the number of missing observations.
The frequencies of the groups defined by <aggr_variable> are
saved in variable 'FREQ'. An alternative name may be given by
specification FREQ=<name_of_variable>.
The IND, CASES and SELECT can be used to select observations.
The type (1,2,4 or 8) of new variables can be given by
TYPE=<type>. Default is TYPE=4.
If the <data> includes variables which are already aggregated, i.e.
they have a constant value within the groups defined by
<aggr_variable>, they may be copied into the <new_data_file> as such,
by activating them with 'C'.
M = More information on data aggregation
Help System (web edition)