LISTSERV at the University of Georgia
Menubar Imagemap
Home Browse Manage Request Manuals Register
Previous messageNext messagePrevious in topicNext in topicPrevious by same authorNext by same authorPrevious page (June 2005)Back to main SPSSX-L pageJoin or leave SPSSX-L (or change settings)ReplyPost a new messageSearchProportional fontNon-proportional font
Date:   Thu, 2 Jun 2005 20:34:42 -0300
Reply-To:   Hector Maletta <hmaletta@fibertel.com.ar>
Sender:   "SPSSX(r) Discussion" <SPSSX-L@LISTSERV.UGA.EDU>
From:   Hector Maletta <hmaletta@fibertel.com.ar>
Subject:   Re: banner tables in SPSS
Comments:   To: Nico Peruzzi <nperuzzi@gmail.com>
In-Reply-To:   <1dbd313405060215547d03b7e7@mail.gmail.com>
Content-Type:   text/plain; charset="US-ASCII"

Nico, I do not know whether your banner tables have anything special, but tables with the shape you describe are extremely easy to obtain in SPSS (with TABLES or CTABLES). The various "mast" questions in the rows, crossed by a single "banner" variable at the top, are just concatenated with the + sign. For instance:

TABLES /PTOTAL=TOTAL /table total+varA+varB+varC+varD by VarE.

To this you may add an OBSERVATION and/or a /STAT subcommand to specify what goes into the cells: absolute frequencies, percentages (by row, column or table), or a summary measure of another variable. To put in each cell the average of income, for instance, the syntax is:

TABLES /PTOTAL=TOTAL/OBSERVATION INCOME /table total+varA+varB+varC+varD by VarE by INCOME.

This produce means because the mean is the default statistic for observation variables in the absence of a /STAT subcomand, but you can specify other summary statistics for the obseration variable/s (sum, standard deviation, number of valid cases, sum percentages etc.) with an appropriate /STAT subcommand as in:

TABLES /PTOTAL=TOTAL/OBSERVATION INCOME /table total+varA+varB+varC+varD by VarE by INCOME /stat sum /stat spct (pct7.2) '% of total income': vara varb varc vard.

This would produce a table containing the sum of income for each cell, and another table whose rows show the distribution of income for each category of the mast variables (VarA to VarD) across categories of the banner variable (VarE).

You can also produce a single table for a number of variables having the same categories, such as several Likert-type questions, with the common categories across the top:

TABLES /PTOTAL=TOTAL /table total+varA+varB+varC+varD BY (LABELS).

Now, all this is pretty elementary and easily done in SPSS, thus perhaps you are speaking of something else. If so, please clarify.

Hector

> -----Original Message----- > From: SPSSX(r) Discussion [mailto:SPSSX-L@LISTSERV.UGA.EDU] > On Behalf Of Nico Peruzzi > Sent: Thursday, June 02, 2005 7:54 PM > To: SPSSX-L@LISTSERV.UGA.EDU > Subject: banner tables in SPSS > > > I know a lot of people who use a program called Wincross to > create banner > tables (i.e., cross-tabbing a single or series of questions > by another > variable - the levels of that variable being the 'banner > points' across the > top of the page) because of limitations in SPSS. Any new > developments I > should know of in SPSS such that SPSS does this well. I'm > using 11, but > considering 13. > > Thanks, Nico > > -- > Nico Peruzzi, Ph.D. >


Back to: Top of message | Previous page | Main SPSSX-L page