Hello,
I have data on a social norms experiment and wish to have a graph that displays the heterogeneity of treatment effects according to the richness of respondents’ social environments (“rich” vs “poor”).
To create the coefficient plots shown, I have written the code given below. However, I would like to standardise the indices and unfortunately after various tests I am at a loss as to how to adapt my code for standardised indices.
To explain the variables used:
If you wish for any further information, please let me know and I will provide it.
(i) Desired type of graph
(ii) My figure: Treatment effect heterogeneity by richness of social environment (only my Behaviour treatment is given below)
(iii) Code (w/simplified control variables)
I have data on a social norms experiment and wish to have a graph that displays the heterogeneity of treatment effects according to the richness of respondents’ social environments (“rich” vs “poor”).
To create the coefficient plots shown, I have written the code given below. However, I would like to standardise the indices and unfortunately after various tests I am at a loss as to how to adapt my code for standardised indices.
To explain the variables used:
- Respondents indicated on a 5-point Likert scale the degree to which they agreed with 2 separate statements indicating “richness” of social environment
- Strongly agree = coded ‘9’ = “very rich”
- Strong disagree = coded ‘13’ = “very poor”
- The values in the remaining range fill the gap.
- Respondents completed the above for each type of extracurricular activity they participate in e.g., course-based, non-profit, sports etc.
- I created an index for overall social environment richness within each activity.
- From this, I created an index aggregating over all activities to indicate a general measure of social environment richness.
- I did a loop to regress run a regression of my main outcome variable (‘Donations’) on each variable, a treatment indicator, and a set of background characteristics.
- To separately analyse treatment effects for ‘rich’ vs ‘poor’ social environments, responses which averaged to strongly agree or agree (if `var'<10.5) with the given statements are taken to indicate a socially rich environment. Responses averaging as neutral or less are considered as a ‘poor’ social environment (if `var'>=10.5).
- I used the user-written command coefplot
- How can heterogeneity of treatment effects be shown for these standardised indices?
If you wish for any further information, please let me know and I will provide it.
(i) Desired type of graph
(ii) My figure: Treatment effect heterogeneity by richness of social environment (only my Behaviour treatment is given below)
(iii) Code (w/simplified control variables)
Code:
set scheme s1color
foreach var in SEIndex_sports SEIndex_nonprof SEIndex_course SEIndex_relg SEIndex_overall {
quietly regress Donation `var' dGroupid2 Gender if `var'<10.5
estimates store `var'
}
coefplot (SEIndex_sports SEIndex_nonprof SEIndex_course SEIndex_relg SEIndex_overall, label(Behaviour)), drop(dGroupid2 Gender _cons) mlabel(cond(@pval<.001, "<0.001" + "***", ///
cond(@pval<.01, string(@pval,"%9.3f") + "**", ///
cond(@pval<.05, string(@pval,"%9.3f") + "*", ///
string(@pval,"%9.3f"))))) ///
note("p-values shown alongside markers" "* p<0.05, ** p<0.01, *** p<0.001") xline(0) title("{bf: Panel A.1:} Rich social environment") name(Figure3BehRichZ, replace)
eststo clear
foreach var in SEIndex_sports SEIndex_nonprof SEIndex_course SEIndex_relg SEIndex_overall {
quietly regress Donation `var' dGroupid2 Gender if `var'>=10.5
estimates store `var'
}
coefplot (SEIndex_sports SEIndex_nonprof SEIndex_course SEIndex_relg SEIndex_overall, label(Behaviour)), drop(dGroupid2 Gender _cons) mlabel(cond(@pval<.001, "<0.001" + "***", ///
cond(@pval<.01, string(@pval,"%9.3f") + "**", ///
cond(@pval<.05, string(@pval,"%9.3f") + "*", ///
string(@pval,"%9.3f"))))) ///
note("p-values shown alongside markers" "* p<0.05, ** p<0.01, *** p<0.001") xline(0) title("{bf: Panel A.2:} Poor social environment") name(Figure3BehPoorZ, replace)
eststo clear
graph combine Figure3BehRichZ Figure3BehPoorZ
Code:
dataex dGroupid2 Donation Gender SEIndex_sports SEIndex_nonprof SEIndex_course SEIndex_relg SEIndex_overall SEIndex_sports_Zz SEIndex_political_Zz SEIndex_nonprof_Zz SEIndex_course_Zz SEIndex_relg_Zz SEIndex_overall_Zz * Example generated by -dataex-. For more info, type help dataex clear input byte dGroupid2 int Donation byte Gender float(SEIndex_sports SEIndex_nonprof SEIndex_course SEIndex_relg SEIndex_overall SEIndex_sports_Zz SEIndex_political_Zz SEIndex_nonprof_Zz SEIndex_course_Zz SEIndex_relg_Zz SEIndex_overall_Zz) 0 50 5 10 . . . 10 -.02870074 . . . . -.02870074 0 15 5 9 . . 9.5 9.25 -.9106002 . . . -.6535326 -.7820664 0 40 5 9 11 . . 10 -.9106002 . 1.1277212 . . .1085605 0 90 6 9 . . . 9 -.9106002 . . . . -.9106002 0 80 6 9 . . . 9 -.9106002 . . . . -.9106002 0 110 5 . . . 10 10 . . . . -.04202402 -.04202402 0 110 5 10.5 . . 10 10.25 .700974 . . . -.04202402 .329475 0 64 6 . 10 . 10 10 . . -.030996576 . -.0872079 -.05910224 0 84 6 . . 11 10.5 10.75 . . . 1.3673325 .4791168 .9232246 0 55 6 9.5 . 9.5 . 9.5 -.4696505 . . -.4725748 . -.4711126 0 84 5 9.5 . . . 9.5 -.4696505 . . . . -.4696505 0 90 5 9.5 . 9.5 9.5 9.5 -.4696505 . . -.4725748 -.6083487 -.516858 0 110 5 . . . . 10 . -.070030846 . . . -.070030846 0 39 6 . . 10 10 10 . . . .14072762 -.0872079 .026759863 0 110 6 . . . . . . . . . . . 0 50 6 9 . 10 . 9.5 -.9106002 . . .14072762 . -.3849363 0 50 5 . . . . . . . . . . . 0 45 6 10 . 10 10.5 10.1 -.02870074 . . .14072762 .4791168 .1970479 0 100 6 . . . . . . . . . . . 0 110 6 10 . . 10 10 .26002425 . . . -.0872079 .08640818 0 10 6 9 9 10.5 11 10.08 -.9106002 -.070030846 -1.1897144 .687457 1.0002576 .11378258 0 55 5 9 . 9 . 9 -.9106002 . . -1.0193042 . -.9649522 0 110 6 11 . 11 . 10.6 1.1419237 . . 1.3673325 . .8560902 0 90 5 9.5 . . . 9.5 -.4696505 . . . . -.4696505 0 30 5 9.5 . . . 9.5 -.4696505 . . . . -.4696505 0 59 5 10 . . . 10 .26002425 . . . . .26002425 0 55 5 . 10 . . 10 . . -.030996576 . . -.030996576 0 20 6 . . . . . . . . . . . 0 80 5 11.5 . . . 11.5 1.5828733 . . . . 1.5828733 0 30 5 10.5 . . . 9.75 .412249 . . . . -.31752425 0 55 5 9 . 10.5 9.5 9.6 -.9106002 . . .687457 -.6535326 -.29222527 0 28 5 10 . 10 10 10 .26002425 . . .14072762 -.0872079 .0931396 0 20 5 9.5 . . . 9.5 -.4696505 . . . . -.4696505 0 0 5 10 . 9 . 9.5 .26002425 . . -1.0193042 . -.37963995 0 20 5 9.5 . . . 9.5 -.4696505 . . . . -.4696505 0 110 5 . . 9 10.5 10.17 . . . -1.0193042 .4791168 .16081707 0 71 5 . 10 . . 10 . . -.030996576 . . -.030996576 0 20 5 9 . . . 9 -.9106002 . . . . -.9106002 0 110 5 9.5 . 9.5 . 9.5 -.4696505 . . -.4725748 . -.4711126 0 110 6 . 10 . . 10 . . -.030996576 . . -.030996576 0 14 5 10 . . . 10 .26002425 . . . . .26002425 0 75 5 11 . 10 . 10.5 1.7193736 . . .14072762 . .9300506 0 50 6 . 10.5 . 10.5 10.5 . . .4989063 . .3887489 .4438276 0 110 6 . 10.5 10 10.5 10.3 . . .4989063 .14072762 .4339329 .3578556 0 60 5 11 . . . 11 .8531986 . . . . .8531986 0 70 5 10 . . . 10 .26002425 . . . . .26002425 0 95 6 9 . . . 9 -.9106002 . . . . -.9106002 0 110 6 11.5 9 11 . 10.13 1.5828733 -1.4779074 -1.1897144 1.1676135 . .02071625 0 10 5 11.5 . 10.5 . 10 2.1603234 . . .687457 . 1.4238902 0 100 6 10 . . . 10 .26002425 . . . . .26002425 0 52 6 . . . . . . . . . . . 0 15 5 10 9 9 . 9.9 .26002425 .4570154 -1.1897144 -1.0193042 . -.06533051 0 55 5 9 . . . 9 -.9106002 . . . . -.9106002 0 82 5 . . . . 10.5 . . . . . .5408265 0 95 6 . . . . . . . . . . . 0 70 6 9 . 11 . 10 -.9106002 . . 1.3007594 . .19507962 0 10 5 . . 9.5 . 9.5 . . . -.4060018 . -.4060018 0 110 6 10.5 10.5 9.5 10.5 10.25 .700974 . .4989063 -.4725748 .4791168 .3016056 0 11 6 9 9 9 9 9 -.9106002 . -1.1897144 -1.0193042 -1.1746734 -1.0683179 0 20 5 10.5 . . . 10.5 .700974 . . . . .700974 0 1 5 10 . . . 10 .26002425 . . . . .26002425 0 110 5 . 10.5 10.5 . 10.3 . -.4238149 .4989063 .687457 . .25418282 0 1 5 9.5 . . . 9.5 -.18092546 . . . . -.18092546 0 90 6 . . . 11 11 . . . . 1.0454415 1.0454415 0 110 6 11 9.5 . . 10 1.1419237 . -.6598114 . . -.027791044 0 110 5 . . . . . . . . . . . 1 55 6 9.5 . 9 . 9.25 -.4696505 . . -1.0193042 . -.7444773 1 110 6 9.5 . 9.5 . 9.5 -.4696505 . . -.4725748 . -.4711126 1 110 6 . . . 9.5 9.5 . . . . -.6083487 -.6083487 1 0 5 11.5 . . . 11.5 1.5828733 . . . . 1.5828733 1 63 6 . 11 . . 11 . . 1.1277212 . . 1.1277212 1 105 6 9.5 10 9 . 9.6 -.18092546 . -.030996576 -1.0193042 . -.4104087 1 0 5 . . . . . . . . . . . 1 110 6 9 . . . 9 -.9106002 . . . . -.9789488 1 110 6 10 . 10 10 10 .26002425 . . .14072762 -.0872079 .10451466 1 110 5 . . 9 . 9 . . . -1.0193042 . -1.0193042 1 110 6 10 . 9.5 9 9.5 .26002425 . . -.4060018 -1.1746734 -.440217 1 110 5 . . . . . . . . . . . 1 90 5 . . 10.5 . 11 . 1.864892 . .687457 . 1.2761745 1 55 6 9.5 . . 9 9.25 -.4696505 . . . -1.1746734 -.822162 1 80 6 . 10 9 10 9.67 . . -.030996576 -1.0193042 -.0872079 -.37916955 1 50 6 . . . . . . . . . . . 1 50 6 . . . . 10.5 . .4570154 . . . .4570154 1 0 5 . . . 10 10 . . . . -.0872079 -.0872079 1 90 6 9 . 10 . 9.5 -.9106002 . . .074154645 . -.4182228 1 0 5 10 . 11 9.5 10.17 -.02870074 . . 1.3007594 -.6083487 -.09589688 1 110 5 11 10.5 . . 10.75 1.1419237 . .5978183 . . .869871 1 55 6 . . . 10.5 10.5 . . . . .4791168 .4791168 1 63 5 9.5 . . . 9.5 -.4696505 . . . . -.4696505 1 33 6 10.5 . . 10.5 10.5 .412249 . . . .5243007 .4682748 1 33 6 10.5 . . 10.5 10.5 .412249 . . . .5243007 .4682748 1 40 5 10.5 . . 11.5 11 .700974 . . . 1.5213984 1.1111863 1 84 5 9.5 . . . 9.5 -.4696505 . . . . -.4696505 1 20 6 11 . . . 11 1.1419237 . . . . 1.1419237 1 110 6 9.5 9.5 . . 9.5 -.4696505 . -.5608995 . . -.515275 1 90 6 10 . . . 10 .26002425 . . . . .26002425 1 29 6 . 9.5 . 10 9.83 . -.070030846 -.6598114 . -.04202402 -.25728875 1 80 6 . 10 11 11.5 10.83 . . -.030996576 1.3007594 1.5213984 .9303871 1 110 5 11 . 10 10 10.3 1.4306487 . . .14072762 -.0872079 .4947228 1 50 6 . . . 11 11 . . . . .9550737 .9550737 end label values Gender Gender_lbl label def Gender_lbl 5 "male", modify label def Gender_lbl 6 "female", modify label values SEIndex_overall SEIndex_overall
