Help!! I have NIBRS data in .txt files, broken into batches. Of the
ten batches, I need three of them to get the variables I need. Each
batch is in the same order (starts and ends with the same incidents).
The listing of incidents (one per row) is enormous, 2007 has just over
5 million incidents.
I am trying to take variables (columns) from one dataset into another.
Cut and paste times out. Any thoughts anyone? Thanks in advance,
Lostcomm
>Help!! I have NIBRS data in .txt files, broken into batches. Of the
>ten batches, I need three of them to get the variables I need. Each
[quoted text clipped - 7 lines]
>
>Lostcomm
That is a bad size to start with, for someone who has never done
much data analysis.
The obvioius approach is to read in each of the three
files separately, into filea, fileb, and filec; and then
MATCH FILEs [... three file specs ...] /BY IDNUM .
I truly hope that there is a row ID, IDNUM (or whatever).
I ordinarily would not expect files of over a couple hundred
cases to be, reliably, in the same order. But those are
probably computer-written.
If the only way to match them is line-by-line, no ID, you
would have to omit the "/BY IDNUM" specification.

Signature
Rich Ulrich