SPSS and Excel calculate breakpoints for quartiles in different ways. This
is problematic for an office of analysts where some use Excel for analysis
and some use SPSS for analysis. We have decided to solve this problem by
reporting out the highest actual value in q1, q2, q3 instead of
interpolating breakpoints. Most of our data is cross-sectional time-series,
where we would want to report distributional statistics (max, 75th
percentile (highest case in Q3), median (and if different the highest case
in q2), 25th percentile (highest case in Q1), and minimum) for each year.
In an ideal world, I would like to create a macro that could be applied
generically to different universes where the number of cases is unknown, as
would be the number of periods of measurement and then output into tables or
ctables which could be written out to excel.
=====================
To manage your subscription to SPSSX-L, send a message to
[hidden email] (not to SPSSX-L), with no body text except the
command. To leave the list, send the command
SIGNOFF SPSSX-L
For a list of commands to manage subscriptions, send the command
INFO REFCARD
Locked
