import excel "C:\table+graph field match countries.xlsx", sheet("Sheet2") cellrange(A1:I27) clear firstrow
renvars, lower // Switch variable names to lower case
list // Check data
// Bring variable names into systematic shape
foreach x of varlist healthwelfare education socialsciencesbusinesslaw ///
artshumanities services sciencemathcomputing ///
agricultureveterinary engineeringmanufacturing {
ren `x' var_`x'
}
list // Check data
// Reshape data
reshape long var_, i(country) j(field_str) string
// Convert proportion into percentage
replace var_ = var_ * 100
// Get field_str variable into numerical format
encode field_str, gen(field1)
label define field1 1 "Agriculture and Veterinary" ///
2 "Arts and Humanities" ///
3 "Education" ///
4 "Engineering and Manufacturing" ///
5 "Health and Welfare" ///
6 "Science, Math, and Computing" ///
7 "Services" ///
8 "Social Sciences, Business, and Law", modify
// Sort by average segregation per field
list field1 var if country == "EU average"
recode field1 (5 = 1 "Health and Welfare") ///
(3 = 2 "Education") ///
(8 = 3 "Social Sciences, Business, and Law") ///
(2 = 4 "Arts and Humanities") ///
(7 = 5 "Services") ///
(6 = 6 "Science, Math, and Computing") ///
(1 = 7 "Agriculture and Veterinary") ///
(4 = 8 "Engineering and Manufacturing") ///
, gen(field)
// Plot
sort field var_
twoway (scatter field var_ if country != "EU average") ///
(scatter field var_ if country == "EU average", connect(l) lwidth(thick)) ///
, ylabel(1/8, val) xscale(alt) ///
xtitle("Percentage women by field of study" "in European countries") ///
ytitle("") note(" " "{it:Source:} European Labor Force Survey 2011, own calculations.", span) ///
legend(order(2) label(2 "Average" "across" "countries") ring(0) pos(1))
