A more current way to reduce a crosstab to a list of matches?

[Art Kendall]

In the 70s, if i had a crosstab with mostly empty cells. I would write the cells out in CROSSTABS
something like

CROSSTABS VARIABLES=CourtCounty DefendantCounty(1,24)
 /TABLES=BY CourtCounty DefendantCounty
 /WRITE = CELLS.
Then I would read the cells back in and SELECT only those cells that did not have a zero count.
I would then edit that in a word processor.

I am guessing that there is a more  current way to produce an output some thing like
It would not necessarily have both the value and the value label.

1 Allegany County:                3 Baltimore City, 5 Calvert County
2 Anne Arundel County:         7 Carrol County
3 Baltimore City                     4 Baltimore County 2 Anne Arundel County 14 Howard County
...

The need for this is not a frequent occurrence  and I could do it the old way if 15 or 20 minutes, but I thought I would check if there is a more current way.  OMS? Tables?

 

[Jon K Peck]

Why not just use AGGREGATE with the crosstab variables as the breaks and output as a new dataset?  The resulting dataset will only have rows for the cells with positive counts.

If something more subtle is needed, please provide more details.


Jon Peck (no "h") aka Kim
Senior Software Engineer, IBM
[hidden email]
phone: [phone]




From:        Art Kendall <[hidden email]>
To:        [hidden email]
Date:        09/08/2014 06:54 AM
Subject:        [SPSSX-L] A more current way to reduce a crosstab to a list of matches?
Sent by:        "SPSSX(r) Discussion" <[hidden email]>

---

In the 70s, if i had a crosstab with mostly empty cells. I would write the
cells out in CROSSTABS
something like

CROSSTABS VARIABLES=CourtCounty DefendantCounty(1,24)
/TABLES=BY CourtCounty DefendantCounty
/WRITE = CELLS.
Then I would read the cells back in and SELECT only those cells that did not
have a zero count.
I would then edit that in a word processor.

I am guessing that there is a more  current way to produce an output some
thing like
It would not necessarily have both the value and the value label.

1 Allegany County:                3 Baltimore City, 5 Calvert County
2 Anne Arundel County:         7 Carrol County
3 Baltimore City                     4 Baltimore County 2 Anne Arundel
County 14 Howard County
...

The need for this is not a frequent occurrence  and I could do it the old
way if 15 or 20 minutes, but I thought I would check if there is a more
current way.  OMS? Tables?





-----
Art Kendall
Social Research Consultants
--
View this message in context: http://spssx-discussion.1045642.n5.nabble.com/A-more-current-way-to-reduce-a-crosstab-to-a-list-of-matches-tp5727176.html
Sent from the SPSSX Discussion mailing list archive at Nabble.com.

=====================
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


===================== 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

[Art Kendall]

duh!

Note to self "Slap your head."

[Barry]

If you want to use the results as an SPSS data set, it's usually better to run an aggregation, followed by filtering, casetovars/varstocases and possibly even a second aggregation (matched back to the first).

I'm also starting to use OMS - my first use was to take an ANOVA table, save as a data set, trim it just to the units, time periods and p-values, and then matching that back to another data set with means and response counts.  The result was a data set indexed by unit and question, with 8 columns of means and response counts by time period, followed by a p-value for the ANOVA for that question-unit combination for the past two waves.

I exported this to Excel, and had something which Excel could use to make handy reports.



"duh!

Note to self "Slap your head."
Art Kendall
Social Research Consultant"