The following is a 'non-destructive' way of solving the problem. I had actually coded up the same thing Clay Helberg posted using AGGREGATE and FLIP but did not like the idea of bumping the working file. Right before I hit the send button I realized this is really a WEIGHTING problem (PARETO gives a sum but not a mean. Aha WEIGHT BY 1/N!!!) Sorry, can't do this ascending!

* To get a set of variables in descending means graphed in a chart, something like the following will work. Ascending ??? * .

COMPUTE @=1. AGGREGATE OUTFILE 'Tmp'/PRESORTED/BREAK @ /N=N. MATCH FILES FILE * / TABLE 'Tmp' / BY @. COMPUTE RecipN=1/N. WEIGHT BY RecipN. GRAPH /PARETO (NOCUM SIMPLE)=SUM( v1 v2 v3 v4 v5 ). Regards, David Marso

snoopytt@aol.com wrote: > > Am trying to find a faster way (either through syntax or some other > method) of ordering variables in bar charts in SPSS (7.5 or 6.4.1). The > default in SPSS is to order them in the order they appear in the file. > I often want to put them in descending or ascending order, but have not > found a fast way to do this. > I am currently removing all the variables-putting them in the omit > window in the chart editor-and them putting them in one at a time > in the order I want. Anyone have a better (faster) way of doing this? > > -------------------==== Posted via Deja News ====----------------------- > http://www.dejanews.com/ Search, Read, Post to Usenet