Re: st: AW: using extended macros to title graphs

[Eric Booth <ebooth@ppri.tamu.edu>]

>>
>>

I couldn't figure out how to get the outer numeric loop that Nick suggested to work with David's example data/code, but a way I got the title to work is with -levelsof- and looping over the "varlabs" rather than the "vars", ex:

************
*!  watch for wrapping issues !*

clear
inp str35 vars str25 varlabs participant_type1 participant_type2 indicator
"children_with_mental_illness_" "Children with MI" .4 .5 1
"communitywide_prevention_init" "Community-wide prevention" .2 .9 0
"differences_between_families_" "Conflict between families " .3 .2 0
"dysfunctional_or_broken_famil" "Dysfunctional " .47 .29 0
"greater_access_to_diversion_o" "Greater access " .2 .5 1
"greater_tolerance_and_cultura" "Greater tolerance " .6 .2 1
"high_levels_of_transience_and" "High levels " .5 .2 0
"holding_juvenile_offenders_mo" "Holding " .4 .4 0
"homeless_youth_face_special_c" "Homeless " .2 .4 0
"immigrant_youth_face_special_" "Immigrant " .2 .3 1
end
*

local tc  dysfunctional_or_broken_famil  greater_tolerance_and_cultura immigrant_youth_face_special_

levelsof varlabs if indicator==1, loc(tc2)


foreach i in `tc2' {
graph hbar (asis) participant_type1 participant_type2  if indicator==1 & "`i'"==varlabs,  ///
nofill   bargap(4.5) blabel(bar, size(medium) position(outside) format(%4.2f))  ///
ytitle(%Responses) title("Causes `i'", size(small))  ///
legend(size(vsmall)) name(g1, replace)   ///
subtitle(, size(small) span bexpand) graphregion(margin(tiny)) plotregion(margin(vsmall))
}
************
Notice that I changed the condition in the -graph hbar- to " if indicator==1 & "`i'"==varlabs".


~ Eric
__
Eric A. Booth
Public Policy Research Institute
Texas A&M University
ebooth@ppri.tamu.edu
Office: +979.845.6754



On Mar 29, 2010, at 11:06 AM, Nick Cox wrote:

> OK. I was trying not to have to read all your code. 
> 
> You have to fix your outer loop so that it loops over numbers. -vars[text]- won't work unless text is the name of a numeric variable or scalar. 
> 
> Nick 
> n.j.cox@durham.ac.uk 
> 
> 
> -----Original Message-----
> From: owner-statalist@hsphsun2.harvard.edu [mailto:owner-statalist@hsphsun2.harvard.edu] On Behalf Of David Souther
> Sent: 29 March 2010 16:58
> To: statalist@hsphsun2.harvard.edu
> Subject: Re: st: AW: using extended macros to title graphs
> 
> Thanks Nick, but this still didn't work for me (this is one of the
> solutions I had tried before posting this question and that I
> mentioned in my reply to Martin Weiss--so I guess I am not too far
> off)
> 
> 
> On Mon, Mar 29, 2010 at 10:48 AM, Nick Cox <n.j.cox@durham.ac.uk> wrote:
>> I guess the problem lies here:
>> 
>> title("Causes   `=varlabs[`vars[`i']']'")
>> 
>> should be
>> 
>> title("Causes   `=varlabs[`=vars[`i']']'")
>> 
>> as the inner and outer macro manipulations are completely independent.
>> Otherwise put,
>> 
>> vars[`i']
>> 
>> is not the name of a local macro.
>> 
>> Nick
>> n.j.cox@durham.ac.uk
>> 
>> David Souther
>> 
>> I am trying to use extended macro functions to title a graph with
>> the value in one string variable that corresponds to another variable.
>> 
>> So, a condensed version of my data looks like:
>> 
>> clear
>> inp str35 vars str25 varlabs participant_type1 participant_type2
>> indicator
>> "children_with_mental_illness_" "Children with MI" .4 .5 1
>> "communitywide_prevention_init" "Community-wide prevention" .2 .9 0
>> "differences_between_families_" "Conflict between families " .3 .2 0
>> "dysfunctional_or_broken_famil" "Dysfunctional " .47 .29 0
>> "greater_access_to_diversion_o" "Greater access " .2 .5 1
>> "greater_tolerance_and_cultura" "Greater tolerance " .6 .2 1
>> "high_levels_of_transience_and" "High levels " .5 .2 0
>> "holding_juvenile_offenders_mo" "Holding " .4 .4 0
>> "homeless_youth_face_special_c" "Homeless " .2 .4 0
>> "immigrant_youth_face_special_" "Immigrant " .2 .3 1
>> end
>> 
>> 
>> I want to graph participant_type(s) for specific conditions (one of
>> the "vars" in `tc' and indicator==1) while adding the corresponding
>> "varlabs" title, here's what I tried (and didn't work):
>> 
>> 
>> local tc  dysfunctional_or_broken_famil  greater_tolerance_and_cultura
>>  immigrant_youth_face_special_
>> foreach i in `tc' {
>> graph hbar (asis) participant_type1 participant_type2  if indicator==1
>> & "`i'"==vars,  ///
>> nofill   bargap(4.5) blabel(bar, size(medium) position(outside)
>> format(%4.2f))  ///
>> ytitle(%Responses) title("Causes   `=varlabs[`vars[`i']']'")  ///
>> legend(size(vsmall)) name(g1, replace)   ///
>> subtitle(, size(small) span bexpand) graphregion(margin(tiny))
>> plotregion(margin(vsmall))
>> *graph save g1 "g1.gph", replace
>> *graph export "causes_`i'_participanttype.pdf", as(pdf) replace name(g1)
>> }
>> 
>> 




*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   http://www.stata.com/support/statalist/faq
*   http://www.ats.ucla.edu/stat/stata/