The study by DMI uses data collected in 1986, before the abolition of the binary divide between traditional universities and former polytechnics. Their results suggest that the earnings penalty associated with attendance at the latter type of institution is about