On Tue, Jul 23, 2013 at 04:30:46PM -0300, Marcelo Brito wrote: > 1. - In cases want to filter using a filter variable, it allows me to select > values for the filter variable. That is: I will "Data" - "Select cases" > and to put the variable in "Use filter variable" allows me to put only the > variable, but not selected to values (could be for example people of 18-25 > years), staying unfiltered base (in the manual is Select if [condition] but > I have no place to enter the condition).
I agree that it would be nice to filter by an expression. As an alternative, you can first compute an expression as a new variable, then filter on that new variable. > 2. - When wanting trivariate analysis, as does the SPSS (I think it's > in contingency > tables or explore), I go to "Analyze" - "Descriptive statistics" - "Browse", > complete the "Dependent List" (in which only allows scalar variables and non > lower-level measurement such as SPSS), "list of factors" and "Label cases by > " (I do not know what it is). The result is a bivariate and trivariate not. I don't know the answer here. _______________________________________________ Pspp-users mailing list [email protected] https://lists.gnu.org/mailman/listinfo/pspp-users
