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 (January 1997)Back to main SPSSX-L pageJoin or leave SPSSX-L (or change settings)ReplyPost a new messageSearchProportional fontNon-proportional font
Date:         Fri, 10 Jan 1997 17:39:07 -0600
Sender:       "SPSSX(r) Discussion" <SPSSX-L@UGA.CC.UGA.EDU>
From:         David Marso <marso@SPSS.COM>
Organization: SPSS Inc
Subject:      Re: Uconventional AGGREGATE?
Content-Type: text/plain; charset=us-ascii

I am having difficulty visualizing the request. Please correct me if I am missing the point here. Suppose one creates a cumulative counter within the variable in question within each normal break. Truncate this cum counter after division by the critical count. then aggregate breaking on the normal breaks and the pseudo-break. *NOTE* this example uses illustrative long names for variables (only 8 characters are permitted in real SPSS code). Hopefully this addresses your question. David Marso

PRESERVE. SORT CASES BY Normal_break_variables Variable_to_be_pseudo_breaked. COMPUTE ONE=1. SPLIT FILE BY normal_break_variables. SET RESULTS OFF MESSAGES OFF. CREATE CCount=CSUM(ONE). RESTORE. COMPUTE Pseudo_Break_Var= Trunc(CCount/Critical_size). AGGREGATE OUTFILE * /PRESORTED / BREAK Normal_break_variables Pseudo_Break_Var / N=N(Variable_to_be_pseudo_breaked).

Finn-Mogens Haug wrote: > > We need to preprocess datasets, aggregating individual > cases into arbitrary groups to reduce the number of "zero > values" in parameters based on counts. Specifically we immuno- > label microscopical objects with gold and calculate areal or > linear particle density to estimate concentrations of > intracellular substances. When labelling is weak, zero particle > counts are frequent. The number of zero-counts can be reduced > by forming (arbitrary) groups of cases and then treat the > groups as if they were individual cases. > > This could be achieved with an imagined option to AGGREGATE, > which would cause a break > > WHEN the aggregated value of a numeric variable > (particle count or number of cases) reaches a limit > OR WHEN the value of some standard break-variable changes > > The scripting language of SPSS does not seem not well suited for > this type of programming. Still, I wonder if someone can point > us to a simple solution for doing the job in SPSS (or SYSTAT). > > Regards > > Finn-Mogens Haug > Department of anatomy,Institute of basic medical sciences, > University of Oslo, N-0317 Oslo, Norway > Tel +47 22 85 12 67 Fax + 47 22 85 12 78 >

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