Contains data from share.dta

obs: $5,854$

vars: $823$ Jan $200915$:$49$

size: $64,394$

Variable storage display

name type format label variable label

$-----------------------------------------------------------------------------------$

school byte $\%8\mathrm{.}0$g school number: $1-25$

sex byte $\%8\mathrm{.}0$g sex: $1=$male, $2=$female

arm byte $\%8\mathrm{.}0$g treatment arm: $0=$control, $1=$intervention

scpar byte $\%8\mathrm{.}0$g social class

debut byte $\%8\mathrm{.}0$g onset of sexual activity in follow$-$up:

$0=$no$,1=$yes

regret byte $\%8\mathrm{.}0$g regret of first sex with last partner $($in

those with $>1$ partner$)$: $0=$no$,1=$yes

kscore byte $\%8\mathrm{.}0$g knowledge of sexual health follow$-$up:

score from $-8$ to $+8$

idno float $\%9\mathrm{.}0$g individual number

$-----------------------------------------------------------------------------------$

Sorted by: school

Sorted by: schCreate a data set containing data for males only from the above dataset and name it "sharem.dta" using stata commands

Use collapse on the data set sharem, collapsing by school and arm, in order to find for each school a$)$ the total number of males in the school $($hint use a $$dummy$$ variable$)$ b$)$ the number of males who have had sexual debut in the school c$)$ the number of males in the school who regret their las sex act. Rename the variables $$male$$debut$\_$m$$ and $$regret$\_$m$$ and create new variables for the proportion of males who had a sexual debut in each school and the proportion of males who regret sex in each school. Save the collapsed data set as $$share$\_$school$\_$m$.$

$11)$ Use summary statistics to compare the school level proportion of males who had sexual debut and the school level proportion of males who regret having sex, between the two study arms.